Reexamination Certificate
1998-12-28
2001-04-10
Patel, Harshad (Department: 2855)
C702S190000
Reexamination Certificate
active
06212975
ABSTRACT:
This invention relates to flow meters, and, in particular, to digital signal processing systems for processing signals from vortex flowmeters.
BACKGROUND
When fluid flows past an obstacle, the obstacle causes a disturbance in the fluid flow. This disturbance is manifested by a vortex generated on one side of the obstacle followed shortly thereafter by another vortex generated on the other side of the obstacle. The two sides of the obstacle continue to alternately generate, or shed, vortices so long as the fluid continues to flow. The frequency at which the two sides of the obstacle shed these vortices is proportional to the velocity of the fluid relative to the obstacle. It is this phenomenon that is the basis for the operation of the known vortex flowmeter.
In a vortex flowmeter, an obstacle in the fluid flow, generally a bluff body, generates an alternating series of vortices. These vortices flow past a pressure transducer at or near the bluff body. Since each vortex is associated with a low pressure zone in the fluid, each time a vortex flows past the pressure transducer, it causes the pressure transducer to generate a pulse having an amplitude proportional to the fluid density and to the square of the fluid velocity. Since the vortices flow with the fluid, the frequency of these pressure pulses is proportional to the fluid velocity. The signal generated by the pressure transducer thus includes a fundamental frequency corresponding to the fluid velocity.
In addition to information about the fluid velocity, the signal generated by the pressure transducer also contains low-frequency components corresponding to other disturbances, such as vibrations from motors, pumps, or unsupported sections of the pipe through which the fluid flows. The transducer signal can also contain high frequency components from other acoustic sources, such as loud noises in the room through which the pipe flows. Additional signal components, both high and low frequency, can also arise from fluid turbulence within the pipe.
These extraneous signal components, collectively referred to as “noise”, are generally filtered out by a bandpass filter having a center frequency at or near the vortex shedding frequency. However, the fact that the vortex shedding frequency is unknown and constantly changing seriously hampers the ease with which one can tune a bandpass filter to that frequency. This difficulty is addressed by the adaptive bandpass filter disclosed in Vignos, U.S. Pat. No. 5,576,497, “Adaptive Filtering for a Vortex Flowmeter,” which is incorporated herein by this reference.
The noise components rejected by the adaptive bandpass filter are not, however, without some value. For example, subtle changes in the spectrum of the noise generated by a pump or motor can foreshadow an imminent mechanical breakdown. Because the pressure signal is responsive to fluid density, changes in the spectrum of the pressure signal can indicate an undesirable change in the composition of the fluid flowing through the pipe. It is therefore useful to monitor the noise components rejected by the adaptive bandpass filter.
In order to adjust the filter passband to match the changing fluid velocity, the adaptive bandpass filter disclosed in Vignos continuously monitors the pressure signal. If the adaptive bandpass filter “looks away,” it is apt to lose track of the fluid velocity and to be unable to recover for some time. As a result, it is impractical to attempt time division multiplexing of the signal from the pressure transducer to the adaptive bandpass filter.
One known approach to observing the noise spectrum is to connect a sweep filter analyzer or similar device directly to the pressure transducer, in parallel with the adaptive bandpass filter. This, however, is a cumbersome procedure since it requires a separate connection at the transducer, an additional piece of hardware, and significant additional power consumption.
What is therefore desirable in the art is a system that can simultaneously track the velocity component of the pressure transducer signal and observe the noise components of that signal.
SUMMARY
The invention provides an integrated digital signal processing system that frequency multiplexes the flowmeter signal from a vortex flowmeter and makes it available to an on-line channel, for tracking the velocity component of the signal, and to an off-line channel, for monitoring the noise components of the pressure transducer signal.
One feature of the invention provides a digital signal processing system, embodied in an integrated circuit, that includes a decimator having two output signals, each of which corresponds to the vortex flowmeter signal sampled at a different sampling rate. These two output signals are provided to two separate channels: an on-line channel that includes an adaptive bandpass filter having a passband responsive to the vortex flowmeter signal, and an off-line channel that includes a tunable bandpass filter having a dynamically selectable passband.
The adaptive bandpass filter of the on-line channel is typically responsive to the fundamental frequency of the flowmeter signal and is typically selected to pass only a narrow band of frequencies around the fundamental frequency.
The tunable bandpass filter typically has a passband that does not depend on the flowmeter signal but is instead controlled by an external source. The external source can sweep the passband of the tunable bandpass filter across a range of frequencies, in which case the tunable bandpass filter functions as a sweep frequency analyzer. Alternatively, the passband can be selected to monitor a fixed range of frequencies in the flowmeter signal in order to closely monitor the operation of a component, such as a pump, associated with the generation of noise in that range of frequencies.
In one non-limiting practice of the invention, the decimator is a multistage decimator having a stage for generating a downsampled signal by downsampling the flowmeter signal to a first sampling frequency and a subsequent stage for upsampling the downsampled signal to a second sampling frequency higher than the first sampling frequency. The subsequent stage can include a sample-and-hold circuit for controlling the sampling frequency corresponding to the two components generated as the output of the decimator.
The off-line channel can include an intermediate frequency filter having a fixed passband, preferably centered at sixty degrees, and software or hardware for shifting the second decimator output such that the frequencies of interest in the second decimator output lie in the fixed passband of the intermediate frequency filter. Such software or hardware can include a mixer for multiplying the second decimator output by a square wave having a user-selectable frequency.
The foregoing system thus enables the simultaneous tracking of the velocity component of a vortex flowmeter signal with the monitoring of the noise components of the same signal. When implemented as an integrated circuit, a digital signal processing system according to the invention provides this functionality with virtually no additional power consumption and without requiring connection of additional hardware to the vortex flowmeter.
REFERENCES:
patent: 5429001 (1995-07-01), Kleven
patent: 5576497 (1996-11-01), Vignos et al.
patent: 5748507 (1998-05-01), Abatzoglou et al.
patent: 5804741 (1998-09-01), Freeman
patent: 5873054 (1999-02-01), Warburton et al.
Foley Hoag & Eliot
Liepmann W. Hugo
Oliver Kevin A.
Patel Harshad
The Foxboro Company
LandOfFree
Adaptive filter with sweep filter analyzer for a vortex... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adaptive filter with sweep filter analyzer for a vortex..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive filter with sweep filter analyzer for a vortex... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2485713