Fluid level sensor without moving parts

Information

  • Patent Grant
  • 6351993
  • Patent Number
    6,351,993
  • Date Filed
    Tuesday, May 9, 2000
    24 years ago
  • Date Issued
    Tuesday, March 5, 2002
    22 years ago
Abstract
An apparatus and method of for measuring fluid levels in a fluid tank (30) utilizing a no moving parts fluid level sensor (20). The apparatus comprises a hollow sensing tube (24) sealed to a pressure sensor (22). The pressure sensor (22) measures air column pressure as a function of hydrostatic pressure within the fluid tanks (30). By modifying the design of the hollow sensing tube (24) to incorporate at least 30% of the volume at its bottom end, the accuracy of the fluid level measurements may be increased by decreasing the meniscus level (34) deviation in the hollow sensing tube (24). Further, by adding a flexible diaphragm (62) to the bottom end that does not allow fluid to enter the hollow sensing tube (24), the accuracy of the fluid level measurements may be further increased by eliminating the meniscus level (34) altogether. Also, design modifications may be made to the hollow sensing tube (24) that ensures that the bottom end of the hollow sensing tube (24) is located at the bottom 30A of the fluid tank (30), or that ensures that fluid slosh effect is dampened, or ensures that the hollow sensing tube (24) is located in the center of the fluid tank (30).
Description




TECHNICAL FIELD




The present invention relates to fluid systems and more particularly an apparatus and method for sensing fluid levels in fluid systems.




BACKGROUND




The majority of automobiles today employ a float arm and thick-film sender (resistance) card to measure fluid levels, and more specifically fuel levels, in a tank. The float arm follows the fluid level in the tank, which changes the resistance across a sender card. Although this method has proven satisfactory for a number of years, it has a number of movable components that are not robust for assembly. Further, the bulkiness of the float arm and thick-film sender card assemblies limits the accuracy of fluid level measurements in certain fluid tank designs because the bottoms of the assemblies do not rest near the bottom of the tank.




Recently, fluid level sensors have focused on fluid level sensors without any moving parts and which measure hydrostatic pressure to determine fluid height. In a “no moving parts sensor”, a pressure sensor is connected to a tube and is open to fluid at the bottom. The air column pressure in the tube will be equal to the hydrostatic pressure caused by the fluid. The height of the fluid can be determined by the pressure reading if the density of the fluid is known. The height that is being measured is the distance between the fluid surface and the meniscus level inside the sensing tube. The meniscus level may change, however, as a function of leakage in the tube/sensor connection, temperature fluctuations, compressibility of the air column in the sensing tube, and/or the tilt angle of the tube. As the meniscus level changes, the distance between the fluid surface and the meniscus level could change.




Also, with current fluid level sensors, the bottom of the fluid tank may expand or contract as a function of fluid weight and tank pressure. As such, a fluid level sensor that is fixed in location in the fluid tank may read more or less depending upon whether the fluid tank is expanding or contracting.




Further, readings on current fluid level sensors may vary due to fluid slosh. As fluid sloshes inside a tank, the height of fluid in various portions of the tank varies. Current fluid level sensors typically average a series of readings to dampen the effect of fluid slosh.




It is thus highly desirable to keep the sensing tube free from fluid or at least lessen the effect of changing meniscus levels in the tube to increase the accuracy of fluid level readings. It is also highly desirable to have the sensing tube measuring fluid levels from the bottom of the fluid tank to increase the accuracy of the measurement of fluid levels. It is also highly desirable to lessen the effect of fluid slosh on fluid level sensor measurements.




SUMMARY OF THE INVENTION




It is an object of the present invention to reduce the effect of temperature fluctuations that vary the meniscus levels in the sensing tube. This object may be accomplished in two ways. First, the shape of the tube may be altered to concentrate the volume of the tube near the bottom. In this way, large temperature fluctuations will cause only slight variation in the meniscus level of the tube. Second, a diaphragm may be added to the end of the tube that transmits hydrostatic fluid pressure into the air column, but does not allow fluid to enter the tube. In this way, the depth of the fluid in the fluid tank is always measured from the bottom of the tube to the fluid surface, greatly increasing the accuracy of the measurement.




It is another object of the present invention to provide a “bottom-sensing” design in the tube that will force the tube to rest approximately on the bottom of the fluid tank. In this way, the measurement of fluid level will be closer to the true height of the fluid even if the fluid tank deforms due to the weight of the fluid or tank pressure/vacuum.




Other objects and advantages of the present invention will become apparent upon considering the following detailed description and appended claims, and upon reference to the accompanying drawings.











BRIEF DESCRIPTION OF THE DRAWINGS





FIG. 1

illustrates a known fluid level sensor with no moving parts showing a meniscus level;





FIG. 2

is a table detailing the worst case meniscus height based on a 10-inch fluid depth of the embodiment of

FIG. 1

;





FIGS. 3-12

illustrate embodiments of the present invention; and





FIGS. 13 and 14

illustrate preferred embodiments of the present invention.











DESCRIPTION OF THE PREFERRED EMBODIMENT(S)




In the following figures, the same reference numerals will be used to identify identical components in the various views.




Referring now to

FIG. 1

, a known “no moving parts sensor”


20


is illustrated. The sensor


20


has a pressure sensor


22


(or transducer) coupled to a sensing tube


24


. The sensing tube


24


, which is hollow and open at the bottom


26


, is sealed to the pressure sensor


22


at the top


28


. The sensing tube is preferably composed of a conductive and corrosive resistant material such as plastic. The pressure sensor


22


is electrically coupled to an engine control module (not shown) which receives and interprets the signals from the pressure sensor


22


concerning fluid levels.




The sensor


20


is placed in a fluid tank


30


such that at least a portion of the sensing tube


24


is submerged in fluid


32


. Preferably, the pressure sensor


22


is not submerged within the fluid


32


.




At any given temperature, the air pressure within the sensing tube


24


as measured by the pressure sensor


22


will be equal to the hydrostatic pressure caused by the fluid


32


. The hydrostatic pressure caused by the fluid


32


increases with depth and is based on the density of the fluid as described by the equation:






p


g


=Ygh






where P


g


is the hydrostatic gage pressure, Y is the fluid density, g is the gravitational acceleration constant, and h is the height/depth of the fluid. Thus the pressure can determine height of the fluid if the density of the fluid is known.




As shown in

FIG. 1

, the height h that is being measured is not the height of the fluid


32


in the fluid tank


30


, but instead is the distance between the fluid surface


32


A and the meniscus level


34


(fluid level) inside the sensing tube


24


. For this reason, the measurement does not reflect the true height of the fluid


32


level.




The height h of the meniscus level


34


is affected by several factors, including leakage in the tube/sensor connection, temperature fluctuations, compressibility of the air column in the sensing tube


34


, and tilt angle.

FIG. 2

illustrates the worst case meniscus heights on the system in

FIG. 1

, where the fluid height is 10 inches and a tube diameter of 0.3 inches. As the diagram indicates, some factors are more significant than others. For example, a leakage in the tube


24


/pressure sensor


22


interface could cause the tube to fill completely with fluid, while a properly sealed tube having a tilt angle of 30 degrees will cause only an additional 0.09 inch meniscus displacement.




Several of these meniscus level factors can be solved quite easily. First, the leakage problem may be solved by using good manufacturing processes to assure a sealed connection. The compressibility of the air column may be eliminated by shifting the calibration down by the amount of the meniscus since this is also a function of fluid height. The tilt angle may be eliminated by better installation techniques.




Temperature fluctuation may affect the meniscus level


34


in two ways. It is well known that gases expand at high temperatures and contract at low temperatures. Thus, when a sensing tube


24


is placed into fluid


32


at a first temperature, for example room temperature, and heated to a second temperature, for example 60 degrees Celsius, the air column in the sensing tube


24


will expand and air will begin to escape out of the bottom of the tube. When the air column cools, the gas will contract and fluid


32


will enter the tube


24


, causing the meniscus level


34


to rise. The Ideal Gas Law (PV=nRT) can be used to calculate how much the air column will contract during cooling. If pressure is kept constant, the amount that the air column contracts is equal to the amount of fluid


32


that gets sucked into the tube


24


.




In the case of automotive fuel systems, the extremes of temperatures anticipated in a fuel tank is between −40 and 60 degrees Celsius (233 to 333 degrees Kelvin). If the fuel tank sees both extremes, it is possible that enough fuel could be sucked into the sensing tube


24


to fill about a maximum of 30% of the total volume of the tube


24


. In other words, the meniscus level


34


may vary by 3 inches in a 10 inch long sensing tube


24


having a 0.3 inch diameter, causing a fuel level reading that may be inaccurate by 30%. It is important to note that the meniscus level


34


will not rise above the 30% maximum as long as the temperature fluctuations remain between −40 and 60 degrees Celsius, regardless of the fluctuations, as any subsequent temperature rises will only force fluid


32


, not air, out of the sensing tube


24


.




Referring now to

FIGS. 3-5

, three embodiments of the present invention are shown that will lessen the impact of temperature fluctuation and the accompanying meniscus level change. In each embodiment, the shape of the sensing tube


24


has been changed to concentrate at least 30% of the total volume of the tube


24


near the bottom. In this way, meniscus levels


34


will vary only slightly at extreme temperature fluctuations.




Referring to

FIG. 3

, one embodiment of the present invention is shown having a bulb


38


which comprises approximately 38% of the volume of the sensing tube


24


. The hole


40


in the bottom


38


A of the bulb


38


is small enough to minimize the effects of tilt angle, and the feet


42


on the bottom are designed to allow a small gap (not shown) between the bottom


38


A and the bottom (shown as


30


A on

FIG. 1

) of the fuel tank


30


. In the worst case scenario as described above, wherein 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the bulb


38


of this embodiment may be increased to further lessen the rise in meniscus level


34


.




Referring to

FIG. 4

, another embodiment of the present invention is shown having a bended portion


46


which comprises approximately 30% of the volume of the sensing tube


24


. In this embodiment, the end


48


of the sensing tube


24


is plugged and a hole


50


is drilled at the bottom so that the hole


50


is as close to the bottom


30


A of the fluid tank


30


so that the fluid meniscus


34


is pushed out in hot temperatures before any air bubbles are pushed out. In the worst case scenario as described above, wherein approximately 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the bend


46


of this embodiment may be increased to further lessen the rise in meniscus level


34


.




Referring to

FIG. 5

, another embodiment of the present invention is shown having a unshaped bended portion


44


which comprises approximately 30% of the volume of the sensing tube


24


. In this embodiment, the u-shaped bended portion


44


takes up less space in the fluid tank


30


as compared with the embodiment in FIG.


4


. In this embodiment, the end


48


of the sensing tube


24


is plugged and a hole


50


is drilled at the bottom so that the hole


50


is as close to the bottom


30


A of the fluid tank


30


so that the fluid meniscus


34


is pushed out in hot temperatures before any air bubbles are pushed out. In the worst case scenario as described above, wherein approximately 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the u-shaped bended portion


44


of this embodiment may be increased to further lessen the rise in meniscus level


34


.





FIG. 6

illustrates another embodiment of the present invention. In this embodiment, a diaphragm


62


is placed on the end of a sensing tube


24


. The diaphragm


62


deflects back and forth with the expanding and contracting air column without allowing fluid to enter the sensing tube


24


. The diaphragm


62


is designed to accommodate the wide range of possible temperature fluctuations (−40 to 60 degrees Celsius) with sufficient expansion/contraction characteristics to resist the buildup of excess pressure/vacuum in the sensing tube. Therefore, it is critical that the diaphragm is “floppy”, rather than “tight”. In addition, the end


64


of the sensing tube


24


has an inverse conical shape that covers the diaphragm


62


and accommodates the 30% changes in volume of the air column while transmitting hydrostatic fluid pressure.





FIG. 7

illustrates another embodiment of the present invention. In this embodiment, a diaphragm


62


is placed on the end of a sensing tube


24


, where the sensing tube


24


has a bended portion


66


similar to FIG.


5


. The diaphragm


62


deflects back and forth with the expanding and contracting air column without allowing fluid to enter the sensing tube


24


. The diaphragm


62


is designed to accommodate the wide range of possible temperature fluctuations (−40 to 60 degrees Celsius) with sufficient expansion/contraction characteristics to resist the buildup of excess pressure/vacuum in the sensing tube. Therefore, it is critical that the diaphragm is “floppy”, rather than “tight”. In addition, the end


66


A of the sensing tube has an inverse conical shape that covers the diaphragm


62


and accommodates the 30% changes in volume of the air column while transmitting hydrostatic fluid pressure. In this embodiment, it will be necessary to trap the diaphragm


62


to keep it on the bottom


30


A of the fluid tank


30


.




The addition of a diaphragm


62


as shown in

FIGS. 6 and 7

eliminates temperature fluctuation as a source of error in fluid level measurements in a no-moving parts fluid level sensor. Thus, since no fluid enters the tube, which in turn changes the hydrostatic pressure, fluid levels may be measured more accurately.




Referring now to

FIGS. 8-10

, three more embodiments are shown which incorporate a “bottom sensing” design in the sensing tube


24


. The “bottom sensing” design assures that the bottom of the tube will always be approximately at the bottom of the tank, which is not always in a fixed location because of manufacturing tolerances and flexure in the fluid tank as it expands and contracts due to fluid weight and tank pressure.




In

FIG. 8

, the hollow sensing tube


24


incorporates a u-shaped bend


70


between the top


28


end that is sealed to the pressure sensor


22


and the bottom. The u-shaped bend


70


acts as a spring to ensure that the bottom of the sensing tube


24


is always at the bottom


30


A of the fluid tank


30


.




Referring now to

FIG. 9

, the hollow sensing tube


24


incorporates a spiral bend


72


between the top


28


end that is sealed to the pressure sensor


22


and the bottom. The spiral bend


72


acts as a spring to ensure that the bottom of the sensing tube


24


is always at the bottom


30


A of the fluid tank


30


.




In

FIG. 10

, a spiral bend


74


is incorporated at the bottom of the sensing tube


24


such that 30% of the volume of the hollow sensing tube is contained within the spiral bend


74


. The spiral bend


74


acts as a spring to ensure that the bottom of the sensing tube


24


is always at the bottom


30


A of the fluid tank


30


. Further, the end


69


of the sensing tube


24


is plugged and a hole


68


is drilled at the bottom so that the hole


68


is as close to the bottom


30


A of the fluid tank


30


so that the meniscus level


34


is pushed out in hot temperatures before any air bubbles are pushed out. In the worst case scenario as described above, wherein approximately 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


(not shown in

FIG. 10

) will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the spiral bend


74


of this embodiment may be increased to further lessen the rise in meniscus level


34


.




Referring now to

FIG. 11

, another embodiment of the present invention is shown, in which the sensing tube


24


is bent such that the end


48


of the sensing tube


24


may be positioned in any location in the fluid tank


30


. Ideally, the bottom


48


is located in a position in the middle of bottom portion


30


A of the fluid tank


30


, a location that will see the least fluid level change as a result of tilt angle or fluid slosh. Again, as above, the bending of the sensing tube


24


will act to lessen the height variation as a result of meniscus changes and low and high temperatures.




Referring now to

FIG. 12

, another embodiment of the present invention is shown, in which an orifice


80


is placed in the sensing tube


24


. The orifice


80


functions to dampen the pressure variation in the sensing tube


24


caused by fluid slosh. The smaller the orifice


80


, the greater the damping effect. More accurate readings of fluid levels will occur because variations in the meniscus level


34


caused by pressure changes due to fluid slosh will be minimized.




It is contemplated that any of the improvements listed above in

FIGS. 2 through 12

may be combined to give a cumulative preferred effect. Referring now to

FIGS. 13 and 14

, two examples illustrating the combination of the learnings of these principles is shown.




In

FIG. 13

, a preferred embodiment of the present invention is shown. In this preferred embodiment, the hollow sensing tube


24


incorporates a u-shaped bend


70


between the top


28


end that is sealed to the pressure sensor


22


and a bulb


38


. The bulb


38


contains at least 38% of the volume of the sensing tube


24


. A hole


40


in the bottom


38


A of the bulb


38


is small enough to minimize the effects of tilt angle, and the feet


42


on the bottom are designed to allow a small gap (not shown) between the bottom


38


A and the bottom


30


A of the fluid tank


30


. In the worst case scenario as described above, wherein 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the bulb


38


of this embodiment may be increased to further lessen the rise in meniscus level


34


. Further, the u-shaped bend


70


acts to ensure that the bulb


38


is located at the bottom


30


A of the fluid tank


30


even as the fluid tanks expands and contracts due to fluid


32


weight and tank


30


pressure. This embodiment is particularly suited for applications in which the fluid tank remains stationary, such as for use in a storage tank, because vibration may cause air to escape from the sensing tube


24


. Vibration acts in a similar manner to temperature fluctuation. The larger the vibration, the larger the meniscus level change due to air escaping from the sensing tube


24


.




In

FIG. 14

, another preferred embodiment of the present invention is shown. In this preferred embodiment, the hollow sensing tube


24


incorporates a spiral bend


74


between the top


28


end that is sealed to the pressure sensor


22


a bulb


38


. The bulb


38


contains at least 38% of the volume of the sensing tube


24


. A hole


40


in the bottom


38


A of the bulb


38


is small enough to minimize the effects of tilt angle, and the feet


42


on the bottom are designed to allow a small gap (not shown) between the bottom


38


A and the bottom


30


A of the fluid tank


30


. In the worst case scenario as described above, wherein 30% of the volume of the sensing tube


24


is filled with fluid


32


, the meniscus level


34


will correspondingly only rise approximately 0.4 inches, as compared with approximately 3 inches in a straight tube as shown in FIG.


1


. It is contemplated that the volume in the bulb


38


of this embodiment may be increased to further lessen the rise in meniscus level


34


. Further, the u-shaped bend


70


acts to ensure that the bulb


38


is located at the bottom


30


A of the fluid tank


30


even as the fluid tanks expands and contracts due to fluid


32


weight and tank


30


pressure. This embodiment is particularly suited for applications in which the fluid tank remains stationary, such as for use in a storage tank, because vibration may cause air to escape from the sensing tube


24


. Vibration acts in a similar manner to temperature fluctuation. The larger the vibration, the larger the meniscus level change due to air escaping from the sensing tube


24


.




In preferred embodiments as depicted in

FIGS. 13A and 14A

, the bulb


38


may be replaced with a diaphragm


62


. The diaphragm


62


deflects back and forth with the expanding and contracting air column without allowing fluid to enter the sensing tube


24


. The diaphragm


62


is designed to accommodate the wide range of possible temperature fluctuations (−40 to 60 degrees Celsius) with sufficient expansion/contraction characteristics to resist the buildup of excess pressure/vacuum in the sensing tube. Therefore, it is critical that the diaphragm is “floppy”, rather than “tight”. In addition, the end


64


of the sensing tube


24


has an inverse conical shape that covers the diaphragm


62


and accommodates the 30% changes in volume of the air column while transmitting hydrostatic fluid pressure. Further, the spiral bend


74


acts to ensure that the diaphragm


62


is located at the bottom


30


A of the fluid tank


30


even as the fluid tank


30


expands and contracts due to fluid


32


weight and fluid tank pressure. These embodiments are especially suited for automotive applications, or other applications in which the fluid tank


30


is not stationary, as the diaphragm


62


prevents air from escaping out of the sensing tube


24


due to vibration, an effect that is similar to temperature fluctuations.




The embodiments as described in

FIGS. 13 and 14

are preferred over the prior art for many reasons. First, they eliminate assembly issues with the long float arm. Second, they eliminate corrosion and wear issues associated with a sender card. Third, they eliminate mechanical wear issues associated with the thick-film resistor cards. Fourth, each preferred embodiment facilitates easier packaging in increasingly complex fluid tank shapes because each preferred embodiment is smaller than the prior art. Finally, the no moving parts fluid level sensor provides continuous signal rather than a stepped signal according to the present technology.




Further, these embodiments incorporate the best features of each of the embodiments of

FIGS. 1 through 12

. First, these preferred embodiments reduce meniscus error caused by temperature fluctuations. Second, there is no serious effect of tilt angle in either embodiment. Third, each embodiment ensures that the bottom of the sensing tube


24


is kept in contact with the bottom


30


A of the fluid tank


30


.




While the invention has been described in terms of embodiments and preferred embodiments, it will be understood, of course, that the invention is not limited thereto since modifications may be made by those skilled in the art, particularly in light of the foregoing teachings.



Claims
  • 1. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; and a hollow sensing tube having a top end and a bottom end and a first volume, wherein said top end is sealed to said pressure sensor, wherein said bottom end having an opening and a second volume; wherein said second volume comprises at least 30% of said first volume and wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having a u-shaped bend.
  • 2. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; and a hollow sensing tube having a top end and a bottom end and a first volume, wherein said top end is sealed to said pressure sensor, wherein said bottom end having an opening and a second volume; wherein said second volume comprises at least 30% of said first volume and wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having a spiral bend.
  • 3. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; and a hollow sensing tube having a top end and a bottom end and a first volume, wherein said top end is sealed to said pressure sensor, wherein said bottom end having an opening and a second volume; wherein said second volume comprises at least 30% of said first volume and wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having an orifice for dampening the effects of fluid slosh.
  • 4. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; and a hollow sensing tube having a top end and a bottom end and a first volume, wherein said top end is sealed to said pressure sensor, wherein said bottom end having an opening and a second volume; wherein said second volume comprises at least 30% of said first volume and wherein said bottom end comprises a bulb portion, said bulb portion having at least one foot, said at least one foot preventing said opening from sealing to a bottom portion of the fluid tank.
  • 5. The sensor of claim 4, wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having a spiral bend.
  • 6. The sensor of claim 4, wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having an orifice for dampening the effects of fluid slosh.
  • 7. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; and a hollow sensing tube having a top end and a bottom end and a first volume, wherein said top end is sealed to said pressure sensor, wherein said bottom end having an opening and a second volume; wherein said second volume comprises at least 30% of said first volume and wherein said bottom portion is bent such that said bottom portion is located adjacent to and substantially parallel to a bottom portion of the fluid tank.
  • 8. The sensor of claim 7, wherein said bottom portion comprises a u-shaped bend.
  • 9. The sensor of claim 7, wherein an end of said bottom portion is located in the center of said bottom portion of said fluid tank.
  • 10. A no moving parts fluid level sensor for use in fluid tanks, the sensor comprising:a pressure sensor for measuring air column pressure; a hollow sensing tube having a top end and a bottom end, wherein said top end sealed to said pressure sensor and said bottom end having an opening; and a diaphragm coupled to said bottom end and covering said opening, said diaphragm being impermeable to fluid in the fluid tank and permeable to air.
  • 11. The sensor of claim 10, wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having a u-shaped bend.
  • 12. The sensor of claim 10, wherein said hollow sensing tube has a middle section between said top end and said bottom end, said middle section having a spiral bend.
  • 13. The sensor of claim 10, wherein said bottom end is bent such that said bottom end is located adjacent to and substantially parallel to a bottom portion of the fluid tank.
  • 14. A method for accurately measuring fluid levels in a fluid tank, the method comprising the steps of:sealing a top end of a hollow sensing tube to a pressure sensor that measures air pressure in said hollow sensing tube; placing said hollow sensing tube in the fluid tank such that a bottom end of said hollow sensing tube is immersed in and fluidically coupled to fluid in the fluid tank when said fluid is at its maximum depth and such that said pressure sensor is not immersed in fluid in the fluid tank when said fluid is at its maximum depth; and sealing a diaphragm around said bottom end of said hollow sensing tube such that said fluid may not enter said hollow sensing tube.
  • 15. A method for accurately measuring fluid levels in a fluid tank, the method comprising the steps of:sealing a top end of a hollow sensing tube to a pressure sensor that measures air pressure in said hollow sensing tube; placing said hollow sensing tube in the fluid tank such that a bottom end of said hollow sensing tube is immersed in and fluidically coupled to fluid in the fluid tank when said fluid is at its maximum depth and such that said pressure sensor is not immersed in fluid in the fluid tank when said fluid is at its maximum depth; and modifying a middle portion of said hollow sensing tube between said top end and said bottom end to bias said bottom end of said hollow sensing tube against a bottom of said fluid tank.
  • 16. A method for accurately measuring fluid levels in a fluid tank, the method comprising the steps of:sealing a top end of a hollow sensing tube to a pressure sensor that measures air pressure in said hollow sensing tube; placing said hollow sensing tube in the fluid tank such that a bottom end of said hollow sensing tube is immersed in and fluidically coupled to fluid in the fluid tank when said fluid is at its maximum depth and such that said pressure sensor is not immersed in fluid in the fluid tank when said fluid is at its maximum depth; and introducing an orifice in said hollow sensing tube that functions to dampen the effect of fluid slosh in the fluid tank.
  • 17. A method for accurately measuring fluid levels in a fluid tank, the method comprising the steps of:sealing a top end of a hollow sensing tube to a pressure sensor that measures air pressure in said hollow sensing tube; placing said hollow sensing tube in the fluid tank such that a bottom end of said hollow sensing tube is immersed in and fluidically coupled to fluid in the fluid tank when said fluid is at its maximum depth and such that said pressure sensor is not immersed in fluid in the fluid tank when said fluid is at its maximum depth, wherein said bottom end of said hollow sensing tube comprises at least approximately 30% of the volume of said hollow sensing tube; and sealing a diaphragm around said bottom end of said hollow sensing tube such that said fluid may not enter said hollow sensing tube.
US Referenced Citations (8)
Number Name Date Kind
1646311 Page Oct 1927 A
3992941 McGoldrick Nov 1976 A
4137781 Andrus Feb 1979 A
4665746 Sheppard May 1987 A
5105662 Marsh et al. Apr 1992 A
5172717 Boyle et al. Dec 1992 A
5802910 Krahn et al. Sep 1998 A
6210577 Garber Apr 2001 B1