FIEI J> OF THE INVENTION

This invention relates generally to rotational motion sensors and more particular!; to gear tooth sensors that utilize magnetic field sensing elements for rotational motion

BACKGROUND OF THE INVENTION

Magnetic field sensors that sense the rotational motion of a target are known. The target can be a hard ferromagnetic (permanent magnet) or a soft ferromagnetic target.

Magnetic field sensors that detect the features of a ferromagnetic gear target belong to a class of sensors known as "gear tooth sensors." Gear tooth sensors are key elements in engine management applications, anti-lock braking systems, transmission systems, and other industrial or automotive systems.

In this type of magnetic field sensor, the magnetic field associated with the target's mechanical profile is sensed by a magnetic field sensing element, sach as Hail element or magnetoresistive (MR) element. In the case where the target is a soft ferromagnetic material a permanent magnet may be positioned near the sensor to provide a magnetic field to fee sensing element. As the target's features pass the sensing element, the magnetic field experienced by the sensing element varies in relation to the target profile. The sensing element provides a signal proportional to the sensed magnetic field. The sensor processes the magnetic field signal to generate an oistput signal that changes state each time the magnetic field signal crosses a threshold. Such m output can be used to provide rotational speed information. Some sensors, referred to as differential sensors, contain at least one pair of sensing elements configured in a differential arrangement to differential magnetic field sensors, the difference between the signals pr ided by the two sensing elements is used to generate a differential magnetic field signal indicative of transitions in the target's features. The differential sensor processes the differential magnetic field signal to generate an output signal that switches states each time the differential magnetic field signal crosses a threshold. Thus, detection of me approach and retreat of each feature of a rotating ferromagnetic gear results in an output signal that is typically a square wave representation of rotation of the ferromagnetic gear.

Various types of threshold-based mechanisms can be nsed to generate the output signal. One such mechanism is referred to as a threshold detector or peak-to-peak percentage detector (and referred to herein as a threshold detector). It uses thresholds defined as a percentage of the peak-to-peak magnetic field signal (e,g., 40% and 60%), One type of threshold detector is described in U.S. Patent No. 5.9! 7,320 entitled

DETECTION OF PASSING MAGNETIC ARTICLES WHILE PERIODICALLY ADAPTING DETECTION THRESHOLD, w ich is assigned to the assignee of the subject application. Each threshold is determined based on a previous negative and positive peak. If the magnetic field signal doesn't cross one of these thresholds (e.g., because of a sudden large signal shift), the thresholds cannot be updated If the thresholds are sot updated, the output signal will stop switching.

Another mechanism, referred to as a slope-activated, peak-referenced or peak detector (and referred to herein as a peak detector), uses a fixed or variable threshold referenced to the last positive or negative peak (i.e., the last peak or valley) of the magnetic field signal. In the peak detector, the threshold differs from the positive and negative peaks of the magnetic field signal by a predetermined amoun In this type of detector, the output signs! changes state when ths magnetic field signal corses sway from a peak or valley by the predetennined amount. One type of peak detector is descri bed in U.S. Patent No. 6,09! ,239 entitled DETECTION OF PASSING MAGNETIC ARTICLES WITH A PEAK-REFERENCED THRESHOLD DETECTOR, which is assigned to the assigaee of the subject application. Another example of a peak detector can be found in U.S. Patent No, 6,693,419 entitled PROXIMITY DETECTOR, also assigned to the assignee of the sisbjeet application.

ϊη order to accurately detect the passing features of the rotating ferromagnetic object, these circuits must, be capable of closely tracking the magnetic field signal.

Typically, one or more digital-to-analog converters (DACs) are used to generate a DAC signal, which tracks fee magnetic, field signal. is some architectures, ne DAC is used, to track the positive peaks of the magnetic field signal (PD AC) and another DAC is used to track the negative peaks of the magnetic field signal (NDAC). In. other architectures, a single DAC tracks both the positive and negative peaks of the magnetic field signal.

Both detection schemes have their advantages and disadvantages. The peak detector tends to offer a more robust approach to switching, with guaranteed switching even in the presence of large signal variation. The peak detector is commonly used for sensors in automotive anti-lock braking systems (ABS) and transmission applications where magnetic anomalies may compromise the functionality of the sensor. The threshold detector, on the other hand, tends to provide greater switching accuracy.

Some gear tooth sensor architectures incorporate both threshold and peak detection in order to differentiate vibration from true rotation. Examples include the ATS651LSH, ATS655/7LSH and ATS692/3LSH family of differential Ha!l-effecl gear tooth sensors available from Allegro Microsystems, inc. SUMMARY OF THE INVENTION

In general, in one as ec , tie invention Is directed to a sensor that meludes: a magnetic field signal generating circuit including at least one sensing element to generate a .magnetic field signal indicative of features of a rotational ferromagnetic object as fee features pass is front of the at least one sensing element, a threshold detector, responsive to the magnetic field signal, to generate a 'threshold detector output signal, and a peak detector, responsive to the magnetic field signal, to generate a peak detector output signal. The threshold and peak detector output signals ach have transitions associated with cycles of the magnetic field signal. Also included in the sensor is a circuit, coupled to the threshold detector and peak detector, to detect transitions of the threshold detector output signal and peak detector output signal, and to use the detected transitions to detect an error if a sequence is which the tr nsitions occur deviates from an expected sequence.

Embodiments of the invention may include one or more of the following features. The circuit can operate to generate an error signal having a level indicative of the error when, such error is detected. The circuit can include an edge detection circuit to detect the transitions of the peak and threshold output signals and an error detection circuit coupled to the edge detection circuit and responsive to the transitions detection to generate the error signal The edge detection circuit can include a first edge detector to detect each transition of the peak detector output signal and a second edge detector to detect each transition of the threshold detector output signal.

The error detection circuit can include a first circuit, responsive to the edge detection circuit, to monitor the sequence of the transitions of the- peak detector output signal and the threshold detector output signal and to generate a first circuit output signal having a level indicative of a first error when the monitoring detects that the sequence has deviated from the expected sequence. The first circuit output signal can be provided as the error signal. The first circuit can include a counter configured to increment a count value in response to each detected peak detector output signal transition and reset, the count value in response to each detected threshold detector output signal transition. The first dram can further include a comparator to compare the count value to a count limit and set the first circuit output signal level to the level indicative of the first error when the count value reaches the count limit The first circuit can include a counter configured to increment a count value in response to each detected ^' threshold detector output signal transition and reset the count value in response to each detected peak detector output signal transition, and can further include a comparator to compare the count value to a count limit and set the first circuit output signal level to the level indicative of the first error when the count value reaches the count limit

The error detection circuit can further include a second circuit, responsive to the detected peak detector output signal transitions, to compare a peak-to-peak value of the magnetic field signal associated with the detected peak detector output signal transitions for a current cycle to an expected peak-to-peak ^' value and to provide a second circuit output signal based on the comparison. The expected peak-to-peak value can be a peak- to-peak value of the magnetic field signal for a corresponding cycle of a prior rotation of the target. The second circuit can include a comparator to compare the peak-to-peak value to a predetermined percentage of the expected peak-to-peak value, The error detection circuit can further include a logic circuit to generate the error signal based on the first circuit output signal and the second circuit output signal. Each of the first and second edge detectors can operate to generate an output signal that provides a pulse for each detected transition.

The sensor can further include an output signal selector having a first input responsive to the threshold detector output signal a second input responsive to She peak detsctor output signal, and a control signal to select one of the threshold and peak detector output signals as a output signal based on. the error signal generated by the circuit The threshold detector, when configured to operate in a first socle followed by a second mode, can be caused to return to operating in ihe first mode when the level of the error signal indicates an error. The threshold detector can operate in the first mode for a recalibration time period. When, the recahbration time period is complete, the threshold detector can resume operating in the second mode and generate a signal to reset the state of the control signal.

The at least one sensing element can include a pair of sensing elements and the magnetic field signal can be a differential magnetic field signal based on differential sensing by the pair of sensing elements. Each of the at least one sensing element can be a selected one of a Hall-effect element, magneto-resistance (MR) element or other type of magnetic field sensitive element. Material used to implement each of the at least one sensing element can be a selected one of type IV semiconductor material or a type M-V semiconductor material. The ferromagnetic object can be a hard ferromagnetic material. The ferromagnetic object can ^' be a soft ferromagnetic material magnetized by a biasing permanent magnet

In another aspect, the invention is directed to a sensor that includes: a magnetic field signal generating circuit includi g at least one sensing element to generate a magnetic field signal indicative of features of a rotational f errom agnetic object as the features pass in front of the at least one sensing element, a threshold detector, responsive to the magnetic field signal, to generate a threshold detector output signal, and a peak detector, responsive to the magnetic field signal, to generate a peak detector output signal. The threshold and peak detector output signals each have transitions associated with cycles of the magnetic field signal. Also included in the sensor is a circuit, coupled to the threshold detector and peak detector, to detect the transitions of the peak detector output sign l and use the detected transitions to detect an error when a peak-to-peak value of the magnetic field signal associated with the detected transitions for a current cycle differs from an expected peak-to-peak value by a predetermined amount.

Embodiments of the invention may include one or more of the following features.

The circuit can operate to generate an error signal having a level indicative of the error when such error is detected. The circuit can include an error detection circuit to compare the peak-to-peak value to a predetermined percentage of the expected a peak-to- peak value and to set the error signal based on. the comparison. The expected peak-to- peak value can be a peak-to-peak value of the magnetic field signal for a corresponding cycle of a prior rotation of the target

The sensor can farther include an output signal selector having a first input responsive to the threshold detector output signal a second input responsive to the peak detector output signal, and a control signal to select one of the threshold and peak detector output signals as an output signal based on the error signal generated by the circuit The threshold detector, when configured to operate in a first mode followed by a second mode, can be caused to return to operating in the first mode when the level of the error signal indicates an error. The threshold detector can operate in the first mode for a reealibratiori time period. When the recalibration time period is complete, the threshold detector can resume operating in the second mode and generate a signal to reset the state of the control signal.

The at least one sensing element can include a pair of sensing elements and the magnetic field signal can be a differential magnetic field signal based on differential sensing by the pair of sensing elements. Each of the at least one sensing element can ^' be a selected one of a Hall-effect element, magnetoreslstance (MR) element or other type of magnetic field sensitive element. Materia! used to implement each of the at least one sensing element can be a selected one of ty e IV semiconductor material or a type III- V semiconductor material. The ferromagnetic object can be a hard ferromagnetic material. The ferromagnetic object can be a soil ferromagnetic material magnetized by a biasing permanent magnet.

The sensor architecture presented herein uses the information of a threshold detector and peak detector operating in parallel to detect anomalies of the magnetic field signal that may lead to a malfunction of the sensor or a reduction in switching accuracy. One feature monitors the output of both detectors to determine if the information is not congruent, for example, one detector switches twice while the other doesn't switch. If that is the case, an anomaly is detected and the information can be used to take some corrective action, for example, re-starting the sensor, sending a diagnostic message to the user/application or adjusting me switching strategy. Additionally, the presence of the peak detector in the architecture allows detection of every peal". Consequently, peak-to-peak information can be used to monitor the behavior of target features over multiple rotations. Information of runout, loofc-to-tooiii variation and other mechanical irregularities can easily be extracted from the operations of the sensor. The sensor can therefore be used to optimize switching accuracy, even with degraded magnetic field signals.

The foregoing features of the invention, as well as the invention itself, may be more fully understood from fee following detailed description of the drawings, in which;

FIG J shows a block diagram of an exemplary differential magnetic field sensor (for ferromagnetic gear tooth motion detection) thai includes threshold and peak detectors, a switching events monitoring control logic block to monitor the outputs of the threshold and peak detectors, as output signal selector and a calibration controller;

FIG, 2 shows a side vie of an exemplary sensing arrangement including a sensor (as shown in FIG. 1) with a biasing magnet and positioned in proximity to a ferromagnetic toothed gear target profile;

FIGS. 3A-B show waveforms depicting a differential magnetic field signal with corresponding peak detector output signal and threshold detector output signal;

FIG. 4 shows a state diagram illustrating behavior of the switching events monitoring control logic block from FIG. 1 according to one exemplary embodiment;

FIG. 5 shows a block diagram of the switching events monitoring control logic block (from FIG. 1) that includes an edge detecting circuit and an error detecting circuit;

FIG. 6 shows a logic circuit diagram of an exemplary embodiment of the edge detecting circuit (from FIG. 5);

FIG. 7 shows a logic circuit diagram of an exemplary embodiment of the error detecting circuit (from FIG. 5);

FIG, 8 shows a logic circuit diagram of an exemplary embodiment of the output signal selector from FIG. 1 ;

FIG. 9 shows a logic circuit diagram of an. exemplary embodiment of the calibration controller from FIG. 1 ; and

FIG. 10 shows a block diagram of an exemplary embodiment of the threshold detector from FIG, 1.

DETAILED DESCRIPTION

Referring to FIG. L a magnetic field sensor ("sensor") 10 for ferromagnetic gear rotational motion detection is shown. The sensor 10 includes a magnetic field signal generator 12, a threshold detector 14, a peak-referenced (or peak) detector 16, a switching events monitoring control logic block (or, simply, control logic) 18, as output signal selector 20 and a calibration controller 22. The magnetic field signal generator 12 includes- t least one differentia! sensing aixangemeat of sensing elements shown as a first sensing element ("SE1") 24a and a second sensing element ("8Β2 ^5, > 24b coupled to a differential amplifier 26 by a first sensing element output 28a and second sensing element output 28b, respectively. The differential amplifier 26 generates at its output a magnetic field signal shown as a differentia! magnetic field {"DIFF ⁵ ) signal 30. The DIFF signal 30 represents the difference in magnetic flux density between the two sensing elements.

The DIFF signal 30 is provided as an. input to both the threshold detector 14 and the peak detector 16, The threshold detector 14 generates a threshold detector output signal shown as a "POSCOMP" signal 32, The peak detector 16 generates a peak detector output signal shown as a "POSCOMPJPK" signal 34. Each of these detector output signals is provided as an input to the control logic 18 and the output signal selector 20. Based on these inputs, the control logic 18 operates to generate an output signal

("ERROR" signal) 36 that indicates whether or not an error has occurred.

By "error" it is meant that at least one of the detectors 14, 16 has provided a result that deviates from an expected result, for example, one detector's output signal fails to transition when such transition is expected or a peak as detected by a detector in a current cycle deviates in time from an expected peak, e.g., by a predetermined amount Such deviations may occur as a result of some anomaly associated with the target, e.g., a change in the direction of ths target's rotation or a mechanical irregularity, such as tooth-to-tooth variation, wobble, run-out and the like. The control logic 18 is designed to detect such deviations and flag them as errors. Its output signal, ERROR signal 36, has a first logic level (e.g., an active high) to indicate that an error was detected and a second logic level to indicate that so error was detected. The operation of the control logic 18 and its err r detection capability will be described in greater detail below, in particular, with reference to FIGS. 4-7.

The ERROR signal 36 and a signal shown as a REABYJFH si nal 37, which is generated by the threshold detector 14 is the illustrated embodiment, are provided as inputs to the calibration controller 22, which generates as outputs two control signals. The two output control signals include a first control signal shown as a "RECALIB ATION" signal 38 and a second control signal shown as a reset threshold ("RESETJ ^' iTF) signal 40, ^' The RESET JTH signal 40 is provided as an input to the threshold detector 14, The SECALIBILiVnON signal 38 is provided to the output signal selector 20. in response to the signals 32, 34 and 38, the output signal selector 20 provides a selected one of detector output signals 32 and 34 as an output signal (OUTPUT) 42 at a sensor output (OUT) 44, Also shows is the figure is a clock generator 46 that generates clock signals (collectively, indicated as CLK 48) for use by various clocked subcircuits of sensor 10.

in an application, when the sensor 10 is positioned is proximity to a target, more particularly, a rotating ferromagnetic article such as a toothed gear, fee sensor output si gnal 42 is indicative of the speed of rotation. This output signal is made available to a user or application, it will be appreciated that the ferromagnetic article or object may be a hard ferromagnetic material or a soft ferromagnetic material

Although fee magnetic field signal generator 12 is illustrated as a single differential channel hnplementation, more than one differential ehansel could be used. Multiple pairs of sensing elements and differential amplifiers would be required to realize multiple differential channels. Alternatively, the magnetic field signal generator 12 may be provided as a single-point implementation, which would require o ly a single sensing element, e.g., SE1. There would be no need for a differential functio (performed by the differential amplifier 26 in FIG, !) _s n the single-point implementation. Some magnetic field signal amplification would most likely still, be required.

An external power supply voltage can be provided to the sensor 10 at a VCC terminal or input 50. The sensor's subcircuits are coupled to the VCC input 50 via a VCC bus 52 and are connected to a ground (GND) terminal 54 throug an internal GND connection 56.

The sensor 10 may be provided in the form of an integrated circuit (IC) containing a semiconductor substrate on which the various circuit elements are formed. The I/O interface of such an IC is indicated in the figure by dashed lines 58. The packaged IC would have at least one pin to correspond to each of: the VCC input or terminal SO, GND terminal 54 and the output (OUT) 44. The sensor or sensor IC 10 can be provided with other features and signals not shown. It will be appreciated that the functionality of the IC, that is, the circuit elements contained within it, can be varied to suit, a particular application.

FIG. 2 shows an exemplary sensing arrangement 60 in which the magnetic field sensor 10 is arranged in a radial sensing position relative to a rotating target profile 62, A direction of motion of the rotating target profile 62 is indicated by arrow 64, The target profile 62 faces the "front ^* of the magnetic field sensor 10. to the depicted arraagement, the magnetic field sensor 10 is a differential sensing device and, as such, iacludes the two sensing elements 24a and 24b for generating a differential signal responsive to change in magnetic field strength at a location relative to the target profile 62.

In FIG. 2, the target profile 62 is defined by an alternating pattern of a raised feature (e.g., a tooth) 66 followed by a gap 68, typically referred to as a valley or recess (or notch). The teeth or other type of raised features can have various shapes based on design, requirements. They need not be square-shaped as shown. A target profile pattern such as the one shown in FIG. 2 may be provided along fee eireimrferesce of the rotating target, e.g., a ferromagnetic object such as a toothed gear wheel, cither radially (Le., along the outer rim or edge of the gear) as shown in FIG. 2, or axiaHy (not shown).

In some sensing arrangements, for example, a sensing arrangement in which features of an object made of a soft ferromagnetic material are to be detected, magnetic biasing may be used. Such an arrangement is shown in FIG. 2, In the illustrated arrangement 60 of FIG. 2, the magnetic field is provided by a biasing permanent magnet 70, The south or north pole of the magnet 70 may be attached to (or positioned near) the sensor 10, for example, to the back of the sensor, as shown. The magnet and sensor may be enclosed together as a subassembly n a case 72. Thai subassembly can. further include a magnetic flux concentrator, for example, a magnetic flux concentrator 74 disposed between the magnet 70 and sensor 10, as shown.

Although not shown in FIG. 1 , an output stage provides to sensor output (OUT) a sensor output signal 42 indicative of detected target profile transitions (such as leading and/or trailing tooth edges). The output stage may be implemented as a totem-pole push- pull or open drain, open collector output configuration. This type of configuration is a typical configuration. Devices wife such an output are sometimes referred to as "three- wire" devices. A three-wire device is shown in FIG. 1. Alternatively, the output stage could be implemented as a current source output structure that provides two levels of current representing two di gital output states of the sensor. In. one embodiment, the output current would be provided on the supply/ground lines and the use of a current source output structure would eliminate the need for the output 44. Devices of mis type with the current output structure are sometimes referred to as "two-wire" devices. Alternatively, the output can be coded in a protocol like Inter-Integrated Circuit (!¾), Serial Peripheral Interface (SPI), Single Edge Nibble Transmission (SENT) or other protocols used is automotive, industrial or consumer applications.

Otter aspects of the magnetic field senso 10, not shown, may be implemented according to known techniques and designs. It will be understood that fee device may contain various other circuits that operate collectively to generate fee sensor output from the magnetic field signals of the sensing elements. For example, an amplifier and other circuitry may be coupled between each sensing element and the differential amplifier to amplify the magnetic field and, optionally, to implement other features, such as dynamic offset cancellation (i.e., chopper stabilization), automatic gain control (AGC) and offset adjustment Alternatively, or in addition, such features may be provided elsewhere, e.g., between the differential amplifier and fee detectors 14, 16.

The magnetic field sensing element 24a (or 24b) in this and other embodiments can be, but is not limited to, a Hall effect element, a magnetoresistance element, or a magnetotensistor. As is known, there are different types of Hall effect elements, for example, a planar Hall element, a vertical Hall element, and a Circular Vertical Hall (CVH) element As is also known, there are different types of magnetoresistance elements, for example, a semiconductor magnetoresistance element such as Indium Antimonide (InSb), a giant magnetoresistance (GMR) element, an anisotropic

magnetoresistance element (AMR), a tmmeling magnetoresistance (TMR) element, and a magnetic tunnel junction (MTJ). The sensing element 24a (or 24b) may include a single element or, alternatively, may include two or more elements arranged in various configurations, e.g., a half bridge or full (Wheatstone) bridge. Depending on the device type and other application requirements, the sensing element 24a (or 24b) may be a device made of a type IV semiconductor material such as Silicon (Si) or Germanium (Ge), or a type IEI-V semiconductor material like Gallium-Arsenide (GaAs) or an mdium compound, e.g., ladium-Antimonide (toSb). The GMR device may be a itpiaa-valve GMR or other GMR type, if separate sensing dies are used, the sensing dies could be made of different technologies, for example, GaAs, Ge, AMR, GMR, TMR, ami others.

As discussed above in the Background, peak detectors such, as peak detector 16 track an in ut signal, that is, the magnetic field signal (e.g., D1 F signal 30, in a differential sensing arrangement) and switch at a threshold level related to peaks of the input signal. Threshold detectors such as threshold detector .14 apply levels with which that input signal is compared to induce switching. These thresholds are defined as a percentage of peak-to-peak amplitude (for example, 85 % and 15 %, or 60 % and 40 %). In. the illustrated embodiment, both, of the detectors 14, 16 are assumed to utilize a tracking-based, relative threshold as opposed to as absolute threshold (e.g., a zero crossing threshold).

In one embodiment, as will be described more fully below, the threshold detector 14 is implemented as a threshold detector that operates in a mode referred to herein as a "rotational threshold mode" (RTM). A threshold detector with this mode maintains a history of "prior" thresholds (thresholds generated for prior cycles or prior cycles of prior revolutions) and uses one or more prior thresholds to establish a threshold for the present cycle. A threshold detector such as threshold detector 14 with RTM may be implemented to include features of the architecture described in U.S. Patent Application No.

12/793,813, filed June 4, 2010 and entitled "Circuits and Methods for Generating a

Threshold Signal Used in a Motion Detector," incorporated herein by reference in its entirety.

FIGS, 3A and 3B show waveforms that illustrate output switching performed by the detectors 14, 16 for a differential magnetic field signal. FIG. 3 A shows waveforms 80 including a differential magnetic field signal waveform 82 corresponding to a mechanical target profile. The positive arid negative peaks (of a given cycle) correspond to edges of a target feature. As discussed above, the sensor 10 is capable of providing digital information that is representative of the mechanical features on the rotating target. The waveforms present the translation of the mechanical profile, through the magnetic profile that it induces, to the digital output signal of the sensor. The illustrated waveforms show a particular output polarity for the orientation of a target and sensor that results Irs the output switching from high (ICC(H) or VOUT(H)) to low as the leading edge of a tooth (a rising mechanical edge, as detected by the sensor), passes the sensor package face. In this configuration, the output swi ches to its low polarity when a tooth is the target feature nearest the package. Jf the direction of rotation is reversed, then fee output polarity inverts.

Waveforms 80 also include a peak detector output signal waveform 84 and a threshold detector output signal waveform 86. The output signal of waveform 84 switches (or transitions from one state to the other) when the signal of waveform 82 crosses a positive-peak referenced threshold at point 88a and then switches again when the signal crosses a negative-peak referenced threshold at point 88b. The signal, of waveform 86 also switches when the signal of waveform 82 crosses thresholds. £n the case of the threshold detector, the thresholds are established as percentage of the peak-to-peak signal For simplicity, FIG. 3A illustrates is 50%, 50% threshold setting. Thus, the signal of waveform 86 switches when the signal of waveform 82 crosses the 50% threshold on the downward slope and the upward slope of the differential magnetic field signal. These points are indicated by reference numerals 89a and 89b. The detector output signal waveforms 84 and 86 illustrate the expected sequence of transitions for the peak detector and threshold detectors when operating in parallel manner. For each half cycle, a transition of the peak detector (e.g., at point 88a) is followed by a transition of the threshold detector (e.g., at point 89a), That transition is followed by a transition of the peak detector (e.g.. at point 88b). The typical expected sequence is made of alternating transitions (or switching events), with peak detector transitions followed by corresponding threshold detector transitions. It should be noted that the transitions polarity for the detector output signals can be defined differently (that is the H-to-L transitions could be L- to~H, and the L-to-H transitions H-to-L). The polarity of the detector output signals can be arbitrarily defined.

FIG. 3B shows an example of an inconsistency between the switching of the threshold detector 14 and the switching of the peak detector 16. This inconsistency can be generated by anomalies on the magnetic profile, for example, caused by a direction change, as shown in this figure. Referring to FIG. 3B, waveforms 90 include a differential magnetic field signal waveform 92 generated in response to a target profile of a target experiencing a direction change. Also shown are waveforms 94 and 96 for peak detector and threshold detector output, signals, respectively. The thresholds are the same as in the waveforms of FIG. 3 A. is this iUustxation, transitions of the peak detector occur at points indicated by 98a, 98b, 98c, 98d _s 98e and 98f. Transitions of the threshold detector occur at points indicated by 99a, 99b and 99c for the same time interval. The direction change, as sensed differentially by the differential magnetic field signal, occurs at around time t _A . It can be seen from the figure that the direction change causes the peak detector output signal to switch at point 98d, but the threshold output signal does not switch since the differential magnetic field signal never reaches the 50 % threshold crossing. Thus, there are two consecutive peak detector transitions at points 98c and 98d that are not followed by corresponding transitions of the threshold detector output, signal.

Even if such inconsistencies do not occur, the operation of the peak detector 16 can be used by the sensor to detect anomalies of this kind as well as others by comparing the actual peak-to-peak values for a current cycle to corresponding prior peak-to-peak values, that is, peak-to-peak values for a corresponding cycle of a previous rotation (or revolution), as will be described. In either case, and referring back to FIG. 1 , the ERROR signal 36 can be used to change fee detection (or switching) strategy, e.g., to enable the threshold detector 14 to change from one mode to another mode, or to cause the signal presented at the sensor's output 44 by the output signal selector 20 to be changed from the threshold detector output signal to peak detector output signal, or take other action,

FIG. 4 shows a state diagram representation 100 of the switching events sequence as illustrated in FIGS. 3A-3B and behavior of the control logic 18 in detecting

inconsistencies. Each state is represented by a circle, and the transitions between states are indicated by directed lines connecting the circles, A two-bit binary number inside each circle identifies the state the circle represents. The rightmost bit corresponds to the threshold detector output signal and the other bit corresponds to the peak detector output signal. A first state (' 11 % "Good 1 ^w ) 102 corresponds to a state of me detector output signals when the peak detector output signal is a logic 1 and the threshold detector output signal is a logic 1. A next state ('01 "Good 2") 104 corresponds to a state of the detector output signals after the peak detector output signal has transitioned to a logic 0. A next state ('00% "Good 3") 106 corresponds to a state of the detector output signals after the threshold detector output signal has transitioned to a logic 0. A next state (' 10% Good 4") 108 corresponds to state of the output signals after the peak detector output signal transitions to a logic 1. When the threshold detector output signal transitions to a logic 1 , the sequence is back at state 1.02 with both detector output signals at a logic 1, as indicated, by an arrow 110. Thus, when the pair of detector output signals transition as expected, the sequence of transitions is: state 1 , state 2, state 3, state 4, state 1 ,.. (all "good" or expected states). If an error or deviation from he expected sequence occurs, for example, after state 2 the peak detector output signal transitions to a logic 1 (as indicated by arrow 112) or after state 4 the peak detector output signal transitions to a. logic 0 (as indicated by arrow 114) , then "error" state ("Error 1 * ^* ) 116 is reached. The error state will result in some remedial action being initiated, as indicated by arrow 118.

FIG. 5 shows a block diagram overview of the control logic 18 in which the functionality of the block 18 is partitioned into two portions or units (or circuits), an edge detection unit (or edge detector) 120 and an error detection unit, (or error detector) 122, according to one exemplary embodiment. The edge detection unit 120 receives as inputs the FOSCOMP signal 32 from threshold detector 14, the POSCOMP _. PK signal 34 from the peak detector 16 and a clock signal shown as clock signal 48a from the clock generator 46 (from FIG. 1). When a clock pulse provided by the clock signal 48a is applied to circuitry of the edge detection unit 120, the edge detection unit 120 produces two signals, a threshold output edge detected signal shown as "PC _.. EDGE" signal 124 and a peak detector output edge detected signal shown as "PC _. PK EDGE" signal 126, The

PCJEDGE signal 124 has a level or provides a pulse indicative of an edge or t nsition being detected for the POSCOMP signal 32. The ^' PC_P _EDGE signal 126 has a level or provides a pulse indicative of an edge or transition being detected for the FOSCOMP _P signal 34. The PC EDGE signal 124 and PCJPKJEDGE signal 126 are provided as inputs to the error detection unit 122, which also receives a clock signal shown as clock signal 48b. In response to inputs 1 4, 126 and 48b, and as will be described in greater detail with respect to FIG. 6, the error detection unit 122 generates the error signal 36.

FIG. 6 shows details of die edge detection unit 120 according to one example embodiment The edge detector 120 includes a first edge detector 130a to detect an edge or transition (or switching event) of the threshold detector's POSCOMP signal 32. The edge detector 120 further includes a second edge detector 130b to detect an edge or transition (or switching event) of the peak detector's POSCOMPJPK signal 34. In the illustrated embodiment, the edge detectors 130a, 130b are implemented with, clocked D flip-flops 132a, 132b and exclusive OR gates (XGRs) 134a, 134b, respectively, Turning first to the edge detector 130a, the D flip-flop 132a receives as inputs the POSCOMP signal 32 and the clock signal 48a. The Q output,, output 1 6a, and the POSCOMP signal 32 are provided to fee XO gate 134a, which generates the PCJEDGE signal 124 as an output. The second edge detector 130b has the same configuration as edge detector 130a but uses the POSCOMPJPK signal 34 as the D input. Thus, the D flip-flop 132b receives as inputs the POSCOMP _J PK signal 34 and the clock signal 48a, e.g., the same clock signal provided to fee D flip-flop 132a. The Q output, output 136b, and the

POSCOMPJPK signal 34 are provided to the XOR gate 134b, which generates the

PCJPKJBDGE signal 126. Both edge detectors operate in the same manner. When die D flip-flop is clocked, fee output. Q is set. to the D input. The XOR output briefly goes to a logic 1 only w en there is a change in the state of the D input. Thus, the PCJEDGE signal 124 is generated to provide a pulse at each edge (or transition) of fee POSCOMP signal 32. Similarly, the PC PK EDGE signal 126 is generated to provide a pulse at each edge (or transition) of the POSCOMPJPK signal 34.

FIG. 7 shows details of the error detection unit 122 according to one example embodiment In fee illustrated embodiment, the error detection unit 122 includes a first circuit (or error detector) 140 and a second circuit (or error detector) 142 coupled to a logic circuit ^' shown as an OR gate 144. The OR gate's inputs are the outputs of detectors 140 and 142. The output of the first error detector 140 is shown as a. first output (or OR gate input) 146 and the output of fee second error detector 142 is shown as second output (or OR. gate input) 148. The OR gate 144 provides at its output fee ERROR signal 36 based on the logical ORing (or other equivalent operation) of inputs 146 and 148. Thai is, when at least one of fee input signals h s a level representing a logic 1 , the EEEDE. signal will also h ve a level representing a logic 1. Otherwise, fee level of fee ERROR signal will represent a logic 0,

in this particular embodiment, fee error detection unit 122 Is designed to detect errors based on the results of both of the error detectors 140, 142. .AB error at the output of only one of the error detectors .140, 142 or at the outputs of both error detectors 140, 142 can result in fee indication of art error at output 36.

For example, and as discussed further below, the unit 122 can use the error detector 140 to detect an "out-of sequence" error when one of the threshold or peak detectors (detectors 14, 16) fails to transition as expected. This behavior was illustrated earlier in the state diagram of FIG. 4. This "missed transition" or "out-of-sequence" type of error can occur for various reasons, such as when a target changes direction. A direction change related error was illustrated in FIG. 3B, which showed a peak detector's output signal transitioning before and after a direction change while fee threshold detector's output signal did not transition immediately following fee direction change. In addition, in the illustrated embodiment, the unit 122 can use the second error detector 142 to detect problems that may not give rise to a "missed transition" on he part of one of the detectors 14 _s 16.

in the illustrative embodiment of FIG. 7, the error detector 140 uses a count-based approach and is implemented with a counter 150 and a comparator 152 that performs as a "greater than or equal to 2" ("> 2") comparison operator. The counter 150 receives as its INCREMENT and RESET inputs the two signals PC_PK_EDGE 126 and PC EDGE 124, respectively, and the clock signal 48b. The counter 150 produces as its output a count value 154, which it provides to the comparator 152. The counter 150 increments its count for each peak detector transition or edge indicated by fee PCJPKJEDGE signal 126 and performs a reset for each threshold detector transition or edge indicated by the PC EDGE signal 124. The comparator output (first output 146) is used to indicate an error when the logic operator result is true, that is, the count value 154 is greater than or equal to 2.

Other implementations to detect an "out-of-sequence" type of error are also possible.

The second error detector 142 takes a different approach, one that does not utilize the threshold detector output instead, the second error detector 142 compares a digital peak-to-peak value of the DIFF signal (shown as DDIFF) during a current cycle of DIFF signal, that is, the DIFF signal generated for a particular input feature, e.g., gear tooth, and compares it to a previous peak-to-peak value (previous rotation) as detected by the peak detector for that same feature, e.g., gear tooth. This approach requires that the sensor store internally in a memory unit, e.g.. a Peak-to-Peak History (for a digital representation of the DIFF signal, "DDIFF" signal) 160, a digital peak-to-peak value for positive and negative peaks detected by the peak detector for each feature (cycle) during each rotation or some desired number of most recent rotations. For example, if M is the desired number of rotations for which peak-to-peak information is to be maintained in memory (where M is the value for the current rotation and M-l is the previous rotation), "N" denotes the feature (e.g., tooth) number and n is the number of bits used to represent the peak-to-peak value as a digital value, then the history could be configured to store M n-bit values, for example, it could use an x M (n bits wide by M n-bit words long) storage register (or buffer memory, or other type of data storage element) for each feature number 1, 2,..., N ^« 1, N. Alternatively, instead of storing peak-to-peak values, the history could store the individual positive and negative peak values. Logic could be provided along with sample (i.e., feature and rotation) selection logic to derive the peak-to-peak ^' value from the peak values as needed. Jh one possible implementation, the peak-to-peak history could include an input that tells the memory when to sample new data and change the output data. in one exemplary embodiment, as illustrated in FIG. 7, the second error detector 142 includes a pair ofn~hit shift registers including a first n~bit shift register 162a eorrespoadiag to a positive (or negative) peak of the currently detected feature as detected and a second n-bit shift register 162b corresponding to a negative (or positive) peak of the currently detected feature as detected during the current rotation. The shift registers 162a, 162b each are configured to receive a digital version of DIFF signal 30 (shown as DDIFF value as 30 ^s ) corresponding to the current feature cycle via the respective shift register input (IN), Bot shift registers receive the PC_PK_EDGE signal 126 as a LOAD control and the clock signal 48b. Output lines for shift registers 162a, 162b are indicated by reference numerals 164a, .164b, respectively. When clocked and enabled to load the input data by the PC. PK EDGE signal 126, shift register 162b is loaded with the current half cycle's DDIFF (positive or negative peak) value. A value of fee previous half cycle, having been previously loaded into this shift register, is output and loaded into shift register 162a. The error detector 142 also includes a summer element 168, comparator (or logic operator) 170 and a multiplier element 172 applied to another input value 174 corresponding to a peak-to-peak value for the same feature cycle as observed duriag the last rotation. This input value is shown as PKPK JiXPECTED 174 and is provided by (or retrieved from) the Peak-to~Peak History 1 0. The outputs of the shift registers 162a, 162b, after the registers have been loaded with the positive and negative peaks for the current cycle, are summed at the summer element 168 to produce a current peak-to-peak value (PKPK PK) for the current cycle, value 176. The PKP _EXPECTED value 174 is multiplied by some predetermined or relative amount (V %, for example, where x = 75, as shown) by multiplier element 172. That result, value 178, and the PKPKJPK value 176 are applied as inputs to the comparator 170. Thus, the comparator 170 determines if the current value 176 is less than x % of the previous (PKP _, EXPECTED) value 174 (that is, the value expected for the feature based on what was observed for it during the previous rotation), if it Is (that is, the comparator output is TRUE), the comparator's output, output 148, indicates an. error. Alternatively, or in addition, the comparator could check if the FKP JPKvakie 176 is larger than the PKP EXPECTED value 174 by some predetermined or relative amount. As another option, the PKPK EXPECTED value 174 couid be applied directly as an input to comparator 170.

As mentioned above, the peak-to-peak history 160 can include storage, for example, a collection of registers or other suitable forms of storage, sized to hold samples associated with M rotations of the target and N samples per rotation. Is some

embodiments, the peak-to-peak history can be designed to hold MxN multi-bit (e.g., n-bit, as shown) peak-to-peak samples of the magnetic field signal 18, i.e., a history of the peak- to-peak samples. The storage can store a respective peak value associated with each positive and negative edge of the POSCOMP PK. signal according to the edge detection of the PC _.. FK _. EDGE signal, hi other embodiments, and as illustrated, the storage can store a respective peak-to-peak value produced from the peak samples for a given eycie of the magnetic field signal (for example, the P PK _P value 176 determined by the summer 168).

Although the internal details of tire history 160 are not shown, it will be understood that the history 160 would also include control logic that enables the history 160 to store n-bit samples as they are available and to manage the retrieval of each appropriate n-bit sample from among the possible stored samples for use as the "PKP _^ EXPECTED" value 174.

One particular reason for maintaining and using the history of peak-to-peak samples in this manner is that, if the peak-to-peak value associated with a particular feature deviates significantly from rotation to rotation, the deviation can be used to indicate an error at the error signal 36,

Although the peak-to-peak history 160 is shown here as part of the error detector 142, it could reside elsewhere in the sensor, for example it could be included as part of the peak detector 16. In fact, all of the memory elements and circuits used to produce the PKPKJPK value 176 and the P PK _EXPECTED value 174 for the error detector's comparison could be part of the peak detector 16.

Other embodiments eoidd of course use one or the other of these error detector blocks by itself. For example, in one alternative embodiment, the edge detection unit 1.20 could include both edge detectors 130a and 130b and the error detection unit 122 could include only block 140. In an alternative embodiment, the edge detection unit 120 could include only edge detector 130b and the error detection, unit 122 could include only block 142. Ja either of these alternative embodiments, the logic circuit 144 could be omitted and the output of the error detec tion u it 122 (whether it is the output of block 140 or the output of block 142) would be provided directly as fee error signal 36.

The second error detector 142 advantageously provides capability for detecting errors even if both threshold and. peak detectors are switching as expected (e.g., by monitoring peak-to-peak amplitude of the DIFF signal over time) or if neither of the threshold detector or peak detectors are switching as expected (which might occur if the direction change occurs near an input signal peak).

Using the error detector 142 in combination with fee error detector 140 is a more robust solution than one based on the first error detector 140 alone, as the combined approach not only monitors for missed switching events (as discussed above) but also monitors the behavior of fee target's features over multiple peaks so that run-out or variations in amplitude for a given input feature can be detected. Such variations could be symptomatic of an anomaly, for example, a direction change, or gear wear or malfunction, as mentioned above.

Although the peak detector 16 can normally be expected to switch when the threshold detector switches, the control logic 18 (FIG. 5) could be implemented to also include an edge counting detector like block 140 (FIG. 7) that detects if the

POSCOMPJPK signal 34 is not switching when the POSCOMP signal 32 switches.

FIG. 8 shows an example implementation of the calibration controller 22 (from FIG, 1). In the illustrated implementation, the controller 22 includes a clocked D flip- flop 80, logic circuit shown as an AND gate (with one inverted input) 182 and an SR. flip-flop 184, The D flip-flop 180 has as its D input the ERROR signal 36 (from FIG. 7), a clock input 48c and Q output 186. The Q output is provided to the inverted input of AND gate 182. The AND gate's other input is the ERROR signal 36. The AND gate output is reset threshold signal (RESET __ΊΉ) output 40. The SR flip-flop 184 receives as its S and R inputs the RESBTJTH signal 40 and the READY_TH control signal 37, As indicated in the figure, and shown earlier in FIG. 1, the READYJfH signal 37 is provided from and the RESETJTH signal 40 is provided to the threshold detector 14. Still referring io FIG. 8 _S fee SR flip-flop 184 produces; at its Q output the RECALIBRATIGN signal 38. Based on the operations of the D flip flop 180 and the AND gate logic 182, the state of the RESET_TH signal 40 changes from a logic 0 to a logic 1 only when the state of the ERROR signal 36 changes from logic 0 to logic 1. When the RESET JIH signal 40 is at a logic 1 and The READYJTH signal 37 is a logic 0, the REC ALTERATION signal 38 is logic 1. When the RESET JIH signal 40 returns to logic 0, the RECAUBRATiON signal 38 remains at a logic 1 until the READYJTH signal 37 transitions from logic 0 to logic 1. The RECAUBRATION signal 38 stays at a logic 0 until the RESET TH signal 40 changes to a logic 1 again. The threshold detector 14 can contain control logic to recalibrate in response to some signal, e.g., power-on signal, or event The recalibration requires that the threshold detector 14, if operating is a mode that utilizes a history of thresholds like the above- described RTM, change to a mode that uses a threshold generated for the current cycle (e.g. , a non-RTM) and continues to operate in that mode until it can rebuild its history of thresholds over the course of some number of successive cycles (or even rotations). In the embodiment described herein, that control logic is responsive to the RESET _TH signal 40 ₃ which causes the threshold detector 14 to reset to the non-RTM and recalibrate in response to RESETjra 40. The READY _TH signal 37 can be generated by the threshold detector 14, as shown in the figures, or by some other sensor circuitry. It can be used by the threshold detector 14 to indicate that, based on some predetermined criteria, e.g., a completion of the threshold detector recalibration, if performed, or completion of some number of cycles of operating in. a non-RTM (if not part of a recalibration procedure), the threshold detector 14 is ready to begin operating in the RTM again. The use of the

RESET TH and READ ¥ ^' TH signals is further described below with reference to FIG. 10.

Turning to FIG, 9, details of as example implementation of the output signal selector 20 (from FIG, I) are shown, in the illustrated example, the output signal selector 20 includes a multiplexer (or MUX) 190 configured with the POSCOMP signal 32 and the POSCOMP_PK signal 34 as ^< ø' and ¹ 1 ' inputs, respectively, the RECALIBRATION signal 38 as a select (SEL) control signal, and the OUTPUT signal 42 (the sensor output, from FIG. 1) as its output, When the RECALIBRATION signal 38 is a logic 0, the output signal selector 20 selects the POSCOMP signal 32 (that is, the threshold detector output signal) as tire sensor output. When RECAIIBRATION signal 38 is a logic 1, the output sigaal selector 20 selects fee jPOSCOMPJPK signal 34 (fee output signal of the peak detector) instead

Thus, the output signal selector 20 selects the peak detector output signal

(POSCOMP _. FX. sigaal) 34 as fee OUTPUT 42 when the ERROR signal 36 indicates an error and selects fee threshold detector output signal (POSCOMP signal) 32 after fee threshold detector 14 has completed a reealibration or is otherwise ready to resume operating in RTM (as indicated by the state of the READYJTH signal 37), Alternatively, the sensor could be designed so thai it continues to use the peak detector output 34, or uses only the threshold detector output 32 regardless of whether or not an error is indicated (hi which case the output signal selector 20 could be eliminated).

As described earlier, fee operation of the error detector 140 (FIG. 7) allows the sensor to identify and react to inconsistencies between fee detector output signals 32 and 34. It does so by deterafeiing if the switching events occur in fee expected order. Many input signal abnormalities, including certain direction change scenarios, can be detected due to a violation of the expected order.

The threshold detector 14 may be provided with functionality that allows it to operate in a first mode using a first threshold that is generated for a current cycle or a second mode, introduced above as "RTM", using a second threshold that is generated using a threshold history. An arcMtecture of this type is described in. the above- referenced U.S. Patent Application No. 12/793,813, which also describes using a multiplexer to select between the first threshold and the second threshold based on fee application of a power-on signal

If fee threshold detector 1.4 is operating in RTM, when it processes a passing gear tooth it may use information acquired when the tooth passed during the gear's preceding revolution. If there is a direction change or some other anomaly that causes a "missed transition" to be detected, the state of the output 146 produced by error detector 140 (FIG, 7) will indicate a error. If the state of the output 146 indicates a missed transition on the part of the threshold detector 14, the threshold detector 14 will apply the wrong historic information to the upcoming teeth. To prevent such a scenario, the RBSBT_TH signal 40 (produced by circuit 22, FIG. 8), along with other logic, as described below with reference to FIG, 10, can be used to "reset" aspects of the threshold detector operation.

FIG, 10 shows details of the threshold detector 14 with such RTM capability, but modified to incorporate features responsive to the above-described error signal generation, in particular, a control logic block 200 that is responsive to at least the RESET JfH signal 40 to change the threshold selection (from RTM to non-RTM based) for a predetermined time interval, for example, the time for completing a recalibration, and then generate the READY TH signal 37,

In the illustrated example, the threshold detector 14 includes a threshold generator 202 to receive the DIFF signal 30 after the DIFF signal 30 has been converted from an analog signal to a digital signal (shown as DDiFF 30 ^* ) by an ana og-to-digital converter (ADC) 204 and to generate a first threshold signal 206 based on the DDIFF signal 30 ^s , The threshold detector 14 further includes a multiplexer (MUX) 208, a threshold selection module 210 and a comparator 212, The first threshold signal 206 is provided as one of two inputs to MUX 208 and is also provided to the threshold selection module 210. The threshold selection module is implemented to generate a second threshold signal 214 from an internal threshold history, for example, according to techniques described in the above- referenced U.S. patent application. As shown, the multiplexer 208 is coupled to receive the first threshold signal 206 and the second threshold signal 214, and configured to generate art output signal 216 as a selected one of the first threshold signal 206 or the second threshold signal 214 under control by a select signal 218 generated by the control logic 200, The output signal 216 and DD1FF 30* are provided as inputs to the comparator 212, which produces the POSCOMP signal 32 based in those inputs. The control logic 200 operates to change the state of the select signal 218 to switch the MUX selection from the second threshold signal 21.4 to the first threshold signal 206 when either the

RESETJTH signal 40 or an externally generated power-on (and or calibrate) signal 220 is set to a logic 1 level The threshold signal selection will remain unchanged for a predetermined time or until the occurrence of a predetermined event, at which point the control logic 200 will change the state of the select signal 218 so that the MUX 208 selects as the threshold signal 216 the second threshold, signal 214 and will set the READYJTH signal 37 to a logic 0 level. For example, for a time shortly after the threshold detector 14 is first, powered on, e.g., during an. initial calibration time period, the MUX 208 selects as the output signal the first threshold signal 206, and thereafter, e.g., during a running mode of operation, the MUX 208 can select as the output signal 216, the second threshold signal 214, This arrangement may be necessary immediately after power up, as there is no threshold history and a threshold from a previous revolution of the gear target is not available. When the logic level of the RESET JITI signal 40 indicates the detection of an error detected by control logic 18, it may be desirable to adjust the threshold signal selection for a predetermined time period that may be the same as the calibration time period or a different time period. The RESETJTH signal 40 could instead be applied to the same external power-on/calibration control logic that controls power up and calibration activities for the entire sensor (and feat generates the power-on/calibrate signal 220) and. used to .re-star½ecalibrate the sensor.

Referring back to FIGS. 8 and 9, setting the READYJTH signal 37 to a logic 1 level will serve to reset the RECA BRATION signal 38 so that fee RECAUBRATION signal 38 controls the MUX 190 to select fee POSCOMP signal 32 as the output signal 42, The sensor architecture d scribed harem provides a simple mechanism to prevent sensor malfunction (is the presence of some magnetic anomaly). Ti uses the outputs of threshold and peak detectors for error detection, and does sot require more complex vibration or direction change detecting functionality.

Although the details of the sensor architecture presented herein focus on

applications with rotating targets having mechanical profile features, such as toothed gear wheels, it will be understood that tbe circuits and techniques of the sensor architecture may be applicable to sensing of other types of targets as well. For example, the target could be implemented to include an object made of a hard ferromagnetic material (thai is, a permanent magnet) such as a ring magnet or other type of multi-pole magnet. Such a target could have a profile with a pattern of magnetic 'features' defined by a sequence of alternating north and south magnetic poles, instead of a pattern of mechanical features such as teeth and valleys, as described above.

All references cited herein are hereby incorporated herein by reference in their entirety.

Having described preferred embodiments of the invention, it will now become apparent to one of ordinary skill in the art that other embodiments incorporating their concepts may be used. It is felt therefore that these embodiments should not be limited to disclosed embodiments, but rather should be limited only by the spirit and scope of the appended claims.