This invention relates to methods and apparatus to detect liquid levels in containers.
Detecting fluid levels is important for a variety of reasons. Marine and aviation applications require accurate measurements of fuel to ensure sufficient supplies to reach intended destinations. Aviation applications are exceptionally important to monitor the fuel levels in multiple tanks to ensure proper balance of levels to impart the least impact on a plane's aerodynamics, which can be significantly affected by changes in a plane's three-dimensional center of gravity.
Another important measurement function is to ascertain the presence of, and amount of, any contaminants in fuel to ensure the safe and proper operation of engines operated with the fuel. Entry of contaminants into an operating engine can lead to severe performance problems and even engine failure. A means to constantly monitor the presence and amount of contaminants, particularly water, is an essential component of any fuel measurement system.
An accurate, reliable and safe method of measuring the amount of liquid in a container is essential. Applications include fuel tanks containing volatile liquids, although the invention described herein can accommodate a wide range of liquids, regardless of their volatility characteristics. Other parameters that must be ascertained with accuracy and consistency are the type of fuel and the contamination content, if any. A further consideration is a need for hardware that meets the EMI, ESD and Interface requirements of a container, such as an aviation fuel tank, in its environment in a safe manner.
Prior radar technology includes methods to scan, lock on and track targets. The basic approach is to transmit a signal that scans targets, perform gating on a receiver and select targets to lock onto and track. Analysis of the received signal can then be used to determine the distance (range) of the target and to perform signature recognition to define the type of target and its characteristics. Combining radar technology with transmission line theory through a shielded cable solves the problems attendant with sensing liquid levels in containers, particularly those used in the aviation field.
In one aspect of the invention, a coaxial probe (tube) is connected to a power source and circuitry designed to provide time domain reflectometric readings to determine the level of, and presence of, impurities in a liquid. Also called Micropower Impulse radar, the system involves sending a very low microwave pulse along a waveguide, a probe. At least part of the pulse is reflected at the fluid being measured. The transmitted and reflected pulse times are measured and used as the basis for calculating the fluid's level in a container.
In another aspect of the invention, the coaxial probe connected to the circuitry enables the apparatus to meet all aircraft intrinsic safety requirements such as low electromagnetic interference sensitivity. The apparatus works exceptionally well in a very noisy environment while creating very little noise out. By using a coaxial probe, the apparatus is hermetically sealed and requires very low voltage to operate.
In a further aspect of the invention, a closed loop configuration is used to scan, track and lock onto transmitted and reflected pulses. The propagation speed of the microwave pulse through a material can be used to ascertain the identity of the material through which the pulse is traveling. The propagation speed is directly related to the material's dielectric constant. Materials with different dielectric constants will result in different propagation speeds. Thus, the apparatus can be used to measure fluid and/or gas contamination. The transit time of the pulse is used to measure the dielectric constant also known as the material's relative permittivity.
A further aspect of the invention is to employ a digital signal processor (DSP) to substantially reduce signal-to-noise ratio and improve signal clarity with a novel algorithm to scan, lock and track fluid levels in a container.
These and other objects will be apparent from a reading of the following summary and detailed description along with a review of the appended drawings.
a is a flow chart of the calculations made to convert reflected signals into measurements of height, weight and temperature of a monitored fluid according to one embodiment of the invention.
Referring to
The output of delay 14 is inverted at a high speed digital logic gate (inverting amplifier) 16, the output of which is buffered by high speed transistor 17. Transistor 17 acts as a switch that turns on when a positive or relatively high voltage, e.g., 3.3 volts, is received and turns off when the voltage is low or ground. In one embodiment, inverter 14 puts out either ground or 3+ volts. The output of transistor 17 is converted to a pulse by a capacitor 18 that functions as an open circuit for the slow speed signal or rising edge of the pulse wave, and as a short circuit for the high speed signal or falling edge. In this manner, capacitor 18 prevents passage of the low speed signal.
Capacitor 18 drives the shielded probe which forms a transmission line driven from the source impedance R7. The high speed edge continues along the transmit line into a probe 22, which in one embodiment is shielded, where the source, characteristic and load impedances are optimized to produce large, clearly delineated reflections of various points in a fuel container such as the container bottom (a) and container top (b).
The propagation time of the wave forms or reflections are directly dependent on the dielectric constant of the material being traversed. For example, the propagation time through air will be faster than through fuel because of the differences between the dielectric constants of air and fuel. Additionally, the propagation time will be affected by the amount of fuel or other fluid in the container. A higher fuel content will lead to a slower return pulse.
The depth and type of liquid and size of the container dictate the minimum propagation time and what frequency should be used for the transmitter and receiver rates. The frequency must allow time for the propagation of the transmit signal to the bottom of the container and return to the receiver. In one embodiment, the maximum frequency is used thus allowing integration of the maximum number of returns for good signal to noise ratio. The 1-3 MHz transmit cycle is constantly repeated in a substantially constant manner.
Turning to the receiver, the clock output is low pass filtered via resistor 30 and added to the dc signal from amplifier 42 so that the combined signal sets the threshold for comparator 28. The amplifier output is inverted similar to the inversion performed by inverter 16 except without the preceding delay. The output is buffered by high speed transistor 29 and converted to a pulse by capacitor 26 that couples the high speed falling edge of the waveform to turn a switch 24 on and off. Switch 24 may be implemented by either a Fet switch 68 or a diode switch 70. The result is a receiver gate signal synchronized to the transmit pulse but shifted in time depending on the dc signal from amplifier 42.
Once switch 24 is closed, voltage sampling begins at probe 22. A charge pump 23 is incorporated into the system to follow the transmitted and reflected signals. The sampling switch is closed for relatively short periods of time <<1 nanosecond) and shifted in picoseconds over about a 200 to about a 500 ms cycle depending on the length of the tube and reflection times. Multiple samples are measured at almost the same time spot to insure a high signal to noise ratio. The charge pump includes a capacitor 60 and a resistor 62. Whether the charge pump is charged is dependent on the voltage of the received reflections. If positive voltages are received, the pump is charged.
By sampling multiple times at the same spot, the relatively high frequency reflection is converted to a low frequency dc voltage signal that accommodates the limitations of the DSP. By reducing the frequency, rather than having to take measurements in nanoseconds or picoseconds, measurements can be made in milliseconds.
With this configuration, sampling can be conducted in the same spot relative to the transmit pulse for a relatively long period of time by being shifted (at a picoseconds rate) each time the switch is turned on over this incrementally increasing time differential. Sampling at other spots is accomplished by incrementally increasing the time gap between adjacent switch operation events. DSP 72 initiates the process of setting the time gaps.
An initial command is sent by receiver gate 75 to an amplifier 42 via an integrate and limit algorithm 90 and a 12 bit pulse width modulator (PWM) 94. The dc voltage passes to resistor 36 and a 2 pole low pass filter 37 that includes capacitors 34 and 40 and resistor 38. The voltage is then buffered by amplifier 42. The output voltage of amplifier 42 passes to a resistor 32 and is summed with the output of resistor 30, the sum of which passes to comparator 28. The first command sent results in the initial voltage into resistor 32 being 0 so that there is no time delay effect on the initial signal to switch 24.
The output of comparator 28, which is inverted, passes to switch 44 and resistor 46. Simultaneously with the execution of the delay produced by delay 14, the transmit line pulse travels to switch 48 and resistor 50. The outputs from resistors 46 and 48 are summed and pass through a 2 pole low pass filter 55 that includes capacitors 52 and 56 and resistor 54. The filtered dc signal is amplified by amplifier 58 and enters DSP 72 via a 12 bit a/d converter 82. The digitized signal is summed 84 with the command from receiver gate control 75 and processed by the integrate and limit algorithm 90.
Referring now to
The signal processing begins at start 110. A timer 112 controls the operation. If the timer is ready, the a/d conversion begins at 114. If not, the system loops back and tries again until timer 112 is ready. Once the a/d conversion is complete, the signal from amplifier 58 (represented as vp) is put through a summation step 116 with the voltage command signal from receiver gate control 75 (represented as vr). The result Vin is multiplied by a constant at step 118 and added at step 120 to the immediately preceding voltage output Vout−1 to produce Vout. The magnitude of the resulting Vout is checked at step 122 to determine its magnitude. If the magnitude is greater than a preselected limit, the Vout is set to the selected limit at step 126. If the Vout is less than the selected limit, is passes to the PWM 94. Vout also loops back and is delayed at step 124 to be added to the next Vin.
The output of algorithm 90 is used to set the width of the PWM 94 DSP output and controls the time the receiver gate is started (opened) and ended (closed). The loop will drive the error to zero and thus track any or all return propagation times from the transmit time to the bottom of the container. In one illustrative embodiment, in scan mode, the receiver gate controller 75 output begins at zero time (transmit) and increases until the fluid level and the bottom of the tank/container are detected. Once these times are known, the mode can be commanded to change to lock and track the propagation time (distance) with respect to a) the top of the fuel level, b) through the fluid between the top of the fluid and the bottom of the tank, and c) the bottom of the container only.
Referring again to
Referring to
Filters 76 and 78 (shown in
Algorithm filter 92 (also shown in
Referring to
Referring now to
While the present invention has been described in connection with several embodiments thereof, it will be apparent to those skilled in the art that many changes and modifications may be made without departing from the true spirit and scope of the present invention. Accordingly, it is intended by the appended claims to cover all such changes and modifications as come within the true spirit and scope of the invention.
This application is a continuation of, and claims priority to, copending U.S. nonprovisional patent application entitled “Scan Lock and Track Fluid Characterization and Level Sensor Apparatus and Method,” having Ser. No. 11/650,841, filed Aug. 1, 2007, which is hereby incorporated herein by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
4596144 | Panton et al. | Jun 1986 | A |
5457990 | Oswald et al. | Oct 1995 | A |
5610611 | McEwan | Mar 1997 | A |
5827985 | Grieger et al. | Oct 1998 | A |
5841666 | Perdue et al. | Nov 1998 | A |
5943908 | Innes et al. | Aug 1999 | A |
5973637 | Perdue et al. | Oct 1999 | A |
6078280 | Perdue et al. | Jun 2000 | A |
6232910 | Bell et al. | May 2001 | B1 |
6281801 | Cherry et al. | Aug 2001 | B1 |
6556511 | Welke | Apr 2003 | B1 |
6559657 | McCarthy et al. | May 2003 | B1 |
6626038 | Carsella et al. | Sep 2003 | B1 |
6644114 | McEwan | Nov 2003 | B1 |
6650280 | Arndt et al. | Nov 2003 | B2 |
6690320 | Benway et al. | Feb 2004 | B2 |
6906662 | Faust et al. | Jun 2005 | B2 |
7068051 | Anderson | Jun 2006 | B2 |
7162922 | Freger et al. | Jan 2007 | B2 |
7446695 | McEwan | Nov 2008 | B2 |
20020101373 | Arndt et al. | Aug 2002 | A1 |
20040007061 | Forgue | Jan 2004 | A1 |
20050192727 | Shostak et al. | Sep 2005 | A1 |
20070081617 | Fudge | Apr 2007 | A1 |
20070204689 | Bostrom | Sep 2007 | A1 |
Number | Date | Country |
---|---|---|
0473082 | Mar 1992 | EP |
WO9619715 | Jun 1996 | WO |
WO9926080 | May 1999 | WO |
WO2009046103 | Apr 2009 | WO |
Number | Date | Country | |
---|---|---|---|
20100139393 A1 | Jun 2010 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 11650841 | Jan 2007 | US |
Child | 12630225 | US |