Data processing: measuring – calibrating – or testing – Testing system – Of mechanical system
Reexamination Certificate
1998-07-28
2002-02-12
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Testing system
Of mechanical system
C073S007000
Reexamination Certificate
active
06347289
ABSTRACT:
TECHNICAL FIELD
This invention relates to speed sensors, and more particularly to a method and an apparatus for determining the occurrence of an in-range failure of a speed sensor.
BACKGROUND ART
Various types of sensors are available that sense the speed of movement of different devices. In particular, sensors exist which sense the rotational speed of various parts of machinery such as motors, generators and gas turbine engines. A common type of rotational speed sensor comprises a magnetic pick-up, fixed in position, which magnetically senses the rotational movement of teeth in proximity to the sensor. The teeth, or similar type of physical protrusions, are an integral part of the rotating component such as an output shaft of a generator. The sensor provides an electrical output signal typically having a plurality of discrete voltage pulses that correlate to the motion of the teeth in proximity to the sensor.
The sensed speed signal is usually signal conditioned to remove noise, and is then input to signal processing circuitry. The circuitry comprises a control system for the machinery, wherein the control system calculates the rotational speed of the machinery from some characteristic of the electrical pulses. For example, the speed may be calculated from the number of pulses occurring in a certain period of time. Also typically utilized in the calculation is the diameter of the rotating component and the number of teeth formed in the rotating component.
The calculated rotational speed is utilized by the signal processing circuitry in providing for dynamic control of the machinery. For example, for a gas turbine engine, the calculated engine speed is used by the control system to vary the amount of fuel provided to the combustor portion of the engine. In turn, the amount of fuel is used to maintain or change the current rotational speed of the engine. Thus, a faulty speed signal can lead to the inaccurate control of the machinery, often with disastrous consequences.
For example, a faulty speed signal associated with a generator can falsely indicate to the control system that the generator is experiencing an acceleration condition. This may lead the control system to incorrectly believe that there is a reduced or “dropped” load on the generator. In response, the control system typically causes the generator to enter a rapid deceleration condition in an attempt to prevent the incorrectly perceived generator overspeed condition. The overspeed is an undesirable and incorrect condition that is attributed by the control system to the machinery. The overspeed condition and subsequent incorrect response are caused by the failure of the speed sensor to provide an accurate signal to the control system.
Due to this requirement of an accurate speed signal, the control circuitry often includes some means for periodically checking the health of the speed sensor, typically by checking the validity of the signal provided by the sensor. There are various known methods for determining the existence of a gross failure of a speed sensor. A “gross” failure typically means a situation where the machinery being monitored is moving or rotating, yet the sensor is providing no signal and, thus, a speed of zero is mistakenly indicated. One known detection method involves the use of a plurality of speed sensors connected in a parallel, redundant manner. In this method, the signals provided by the sensors are all compared to one another. A gross failure of a sensor is determined to exist when the sensor output significantly differs from the output of all other sensors.
Another failure situation is where the machinery is rotating either at a constant speed other than zero or at a changing speed (e.g., an acceleration or deceleration). In this situation, the speed sensor is providing the control system with a signal indicative of some speed value other than zero. However, the speed value indicated by the signal is not accurate. For example, the speed of the generator may be constant, yet the sensor indicates that the generator is accelerating or decelerating, or is at a different constant speed value. This situation is commonly referred to as an “in-range” failure of the speed sensor. In-range failures are generally more difficult to determine than gross failures. An in-range failure may be determined by the control system using the aforementioned comparison method employing a plurality of sensors connected in parallel. However, this method is unavailable when only a single speed sensor is used.
DISCLOSURE OF INVENTION
A primary object of the present invention is to provide a method and an apparatus for determining the occurrence of an in-range failure of a speed sensor when only a single sensor is utilized to sense the speed of a moving device such as a rotating piece of machinery.
Another object of the present invention is to provide a method and an apparatus for determining the occurrence of an in-range failure of a speed sensor wherein the method and apparatus are easily implemented in programmable logic.
The present invention is predicated on the fact that when a speed sensor fails in-range, the sensor typically provides an output signal having one or more minimum and maximum peak values. When viewed graphically, these peak values are sharp in nature in that they are relatively quickly attained and then departed from. Also, the speeds indicated by the peak values are in-range, yet the values for the points before and after these peaks may be different. The difference between these values is referred to as the relative change in point. For example, the speed signal may transition from a higher value down to a first low peak value, then it may transition quickly up to a higher peak value, then it may transition quickly back down to a second low peak value, then back up to a higher value, etc. The signal make take on a relatively high frequency sawtooth or triangular waveform configuration. Also, for this example, the first and second low peak values may be different. In contrast, a speed sensor operating normally may indicate an acceleration from a constant speed value by a relatively smooth (i.e., slow) transition from a low value up to a second higher value and without an immediate corresponding transition back down to a lower speed value (i.e., without a deceleration).
According to the present invention, a method and an apparatus for determining the existence of an in-range failure of a speed sensor involves sensing the speed of a moving device at three consecutive instances in time, wherein the corresponding time intervals between these instances in time are preferably equally spaced, and wherein each time interval is relatively short in duration to thereby correlate in time to the waveform characteristics of most anticipated in-range speed sensor failures. The sensed speed at each of these three instances in time is utilized in an excursion formula, preferably implemented by programmable logic, that calculates the total sum of the change in speed between the three instances in time. The excursion formula is defined to be the difference between the sensed speed at the first most recent instance in time and the sensed speed at the second most recent instance in time, with this difference being added to the difference between the sensed speed at the first most recent instance in time and the sensed speed at the current instance in time. The absolute value of this sum is the resulting output of the excursion formula. The calculated total sum of the change in speed is then compared against a predetermined reference value and if the calculated sum exceeds the reference value, an in-range failure of the speed sensor is determined to exist and corrective action may then be taken. With the passage of time and the attainment of the next instance in time, the excursion formula may then be re-applied to the speed values at the then current instance in time, the then first most recent instance in time, and the then second most recent instance in time. In this way, the excursion formula may repeatedly be appli
Hoff Marc S.
Raymond Edward
United Technologies Corporation
LandOfFree
Method and apparatus for determining an in-range failure of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for determining an in-range failure of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for determining an in-range failure of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2964964