Pressure isolated fiber optic torque sensor

Abstract

Fabry-Perot and Bragg grating optical measuring principles are combined with a torsional stress sensing mechanism that converts torque applied in one fluid environment to force exerted in a second environment to measure extreme environmental parameters such as pressure in a petroleum producing borehole.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to a fiber optic type torque sensor for measuring precision values of torque generated within a pressure media such as in an oil or gas well environment. Specifically, this environment will also generally include exposure to high temperature, high pressure, corrosive media, shock and vibration. Additional requirements usually include a small diametrical size, low power consumption and the ability to make accurate measurements in the presence of all of these factors.

2. Description of Related Art

Torque is often measured by the utilization of fiber optic strain gauges in various configurations. These types of measurement techniques, however, are generally limited to values of torque that are high enough to create measurable strain levels within a shaft or torsion element. Also, these configurations would normally only lend themselves to physical configurations that preclude routing of associated optical fibers within a fluid media. These criteria are often not met when measurements are to be made in hostile environments such as below the surface, as in an oil or gas well. Additionally, torque output responses derived from physical measurements often require that the torque should be measured primarily as a force rather than as a displacement. It is an object of this invention, then, to provide a strain gauge type torque sensor, suitable for use with precision physical measurement devices which develop a torque parameter output within such hostile environments.

SUMMARY OF THE INVENTION

The present torque measurement system is comprised of a frictionless pressure isolator to couple torque from a well fluid environment into an instrument environment. A fiber optic based torque displacement, or strain, sensor allows measurement of stress that is imposed in a hostile environment from within the isolated instrument environment. The input torque is transmitted by means of a shaft, immersed within the well pressure media, to a pressure isolator tube. Torque transmitted by the pressure isolator tube is then coupled into a torque-to-displacement converter to generate a displacement, or force that is applied to a fiber optic sensor. Conversely, the torque isolator may be reversed so that the input is via the torque tube and the output is via the torque shaft. In both examples, this approach, as opposed to the more conventional measurement of shear or bending stress, allows very small values of torque, which may be present in a high pressure environment, to be accurately measured. This is also accomplished with a very low resultant input torque displacement response.

Two embodiments of the design are described. with different advantages for each. Both embodiments share common design features to allow them to reject external vibration and provide isolation from coupling to the external support housing with different advantages for each. Also, both share common design features to allow them to reject external vibration and provide isolation from coupling to the external support housing.

BRIEF DESCRIPTION OF THE DRAWINGS

The advantages and further features of the invention will be readily appreciated by those of ordinary skill in the art as the same becomes better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings in which like reference characters designate like or similar elements throughout.

FIG. 1 is an illustration of a pressure isolation tube assembly;

FIG. 2 is an isometric view of a force converter linkage;

FIG. 3 a is a side view of FIG. 2 with displacement conversions;

FIG. 3 b is another side view of FIG. 2 with alternate displacement conversions;

FIG. 4 is an illustration of four force linkages mounted to a torque-to-force converter;

FIG. 5 a illustrates the mounting of additional cross beams to the four force linkages;

FIG. 5 b illustrates how the gap between the added beams responds to an applied torque;

FIG. 6 illustrates an optical fiber coupled to a Fabry Perot cavity formed between the torque responsive beams to measure the differential gap;

FIG. 7 is an illustration of the torque sensor with external pressure on the pressure isolator tube.

FIG. 8 a illustrates two optical fibers mounted within a capillary tube and configured to form a Fabry Perot cavity type load cell;

FIG. 8 b illustrates a single fiber, with a Bragg grating written into it, mounted within a capillary tube to form a load cell;

FIG. 9 is an illustration of the principle of the torque sensor;

FIG. 10 illustrates a mechanical resonator;

FIG. 11 illustrates a another embodiment of the invention;

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The fundamental measurement method conceived for this invention is to develop a torque output, within the sensor, which is representative of the measurand. The term “stress”, as used herein, describes a directed force. “Strain”, as used herein, describes a value of material distortion or stretching due to an applied “stress”. “Torque”, as used herein, describes a specialized “stress” value wherein a force is applied arcuately about an axis to produce a twisted strain.

It is also useful to recall that all materials are resilient to one degree or another. The relationship between stress (force) and strain (distortion) is a characteristic function of material property. Hence, when the operative properties of a material are known, it is possible to determine the magnitude of a stress on that material by measuring the magnitude of strain induced by the stress. In most cases, the converse of this principle is also true.

Torque can be passed from the high pressure environment into an instrument environment by means of a torque-pressure isolation tube as depicted in FIG. 1. In this figure, the isolation tube 10 is simply a cylindrical tube 12 with one end closed at 14 and the other end attached at 16 to pressure vessel 18. The closed end 14 has one end 22 of a shaft 20 attached to it so that an applied torque T₁, less the torque required to deflect the torque tube, will then be applied to the free end 24. It is apparent that the applied torque T₁ (which exists within pressure environment P₁), will result in an output torque T₂ (which exists within pressure environment P₂) and that the input and output torques can be exchanged. Exchanging T₁ and T₂ can be used to reverse the pressure environments, P₁ and P₂.

Of course, the stress effects of a pressure difference between P₁ and P₂ will cause distortions in the pressure isolation tube 10. However, none of these distortions will cause a significant error in the torque values T₁ or T₂ as long as torsional displacements are kept small with respect to the isolation tube 10 effective length. The effect of this arrangement, then, is to allow a torque to be coupled from one pressure level to another without the friction effects of a seal, nor any of the errors associated with distortions caused by the pressure difference.

It is important to note that there will be a torque difference between torque, T₁, and torque, T₂, if any angular displacement is involved. This will occur because of the torque required to produce that displacement within the torque isolation tube, 10. If this angular displacement is kept low then the loss across the pressure isolation tube will also be low. This, then, requires a low displacement input torque sensor to be employed to make the torque measurement.

Strain gauges, as the name implies, are devices that respond to strain. Strain results from the application of a force on a body and its level is dependent on the magnitude of the applied force, as well as the material characteristics and physical dimensions of the body acted upon by the force. This force, in the case of a low level torque, is also low and will generally also result in corresponding low levels of strain. This, then, will generally result in strain gauge responses which may be small with respect to other error effects, such as temperature sensitivities or instrumentation inaccuracies. An objective, then, for measuring low level torque is to be able to increase the strain levels related to the measurement.

Conventional methods of measuring torque with strain gauges do not meet these criteria without also requiring a relatively large rotational displacement. One such approach, for example, is to mount the strain gauges on a rectangular beam which is axially subjected to the torque to be measured. If the beam is made thin then a relatively large twist is required to get significant strains. This can be largely remedied by making the beam thick but, then, relatively large levels of torque are required to produce the desired output response. The result is that this approach does not lend itself to those measurements which simultaneously require sensitivity to low torque and low displacement.

Consider the analogy of stress (force) to volts and displacement to current. Voltage may be accurately measured through a resistance if the current is very small. This would mean that the voltage drop across the resistance due to the small current should be negligible with respect to the voltage to be measured. In the same way, force may be measured accurately, even when discrepancies may be non-linear or otherwise error-prone if the force losses required to drive consequential displacements are small.

FIGS. 2 and 3, illustrate a basic force converter linkage 30, used by this invention, to convert torque to a linear force or displacement D. As shown in FIG. 2, this linkage is a hat section formed symmetrically about its center line 32. This means that angles a and b formed by the junction of the flange panels and the leg panels are equal as are lengths l of leg panels 34 and 36. FIGS. 3a and 3b, illustrate that opposing forces F_C or F_t (or displacements) will result is a force D_O or D_l (or displacement) which is orthogonal to the top crown panel 33 of the force linkage 30

FIG. 4, depicts a top view of four of these force linkages, 30₁ thru 30₄ configured as part of a torque-to-force converter 40. Each linkage 30 is attached with one flange panel mounted to a fixed plate, 42, and the other flange mounted to a central plate 44. Central plate 44 is also attached to torque input shaft 46 and is free to rotate with the shaft about axis 48. These four linkages 30₁ through 30₄ are normally mounted at right angles with respect to each other. It should be apparent that an applied torque on input shaft 46 will result in vertical displacements D_O or D_l of the respective crown panels 33 of the force linkages 30₁ thru 30₄ (FIGS. 2, 3a and 3b). As an example, if the input shaft 46 were to be rotated in a clockwise direction R_CW, then this would also rotate the center plate 44 in a clockwise direction R_CW. Force linkages 30₁ and 30₃ would both have opposing outward displacements F_t applied to them as shown by FIG. 3a while linkages 30₂ and 30₄ would have opposing inward displacements F_c applied as shown by FIG. 3b. Referring to FIG. 3a, crown panels 33₁ and 33₃ respective to linkages 30₁ and 30₃ would have a resulting downward deflection D_l. Also, as shown in FIG. 3b, linkages 30₂ and 30₄ would have their crown panels 33₂ and 33₄ deflected upwards, D_O. The difference between these deflections is proportional to the input torque displacement and this will essentially be independent of other outside factors, such as temperature.

FIG. 5 a illustrates how two beams, 50 and 52, are added and attached to the four linkages, 30₁ thru 30₄. Beam, 50, is attached to the tops of linkages, 30₁ and 30₃, while beam, 52, is mounted on top of linkages 30₂ and 30₄. Beam 50 is disposed to overlay beam 52 with a diametric traverse between the two in the in the vicinity of the axis 48 of torque input shaft 46. A normal traversal angle between the two beams 50 and 52 is not necessary. FIG. 5b shows an end view of beam 50. For the cited example of a clockwise rotation, beam, 50 would have an downward deflection D_l while beam 52 would deflect in the opposite direction D_O. The gap, d, then, would decrease proportionally to the applied torque displacement. Of course, an increase in the gap, d, would occur for an opposite direction torque displacement.

A major advantage of this arrangement is that the distance “d” of FIG. 5b is essentially independent of any small axial deflection of shaft 46, with respect to fixed plate 42 in FIG. 4. This is important because shaft 47 may be displaced by end pressure loading on the isolation tube 10 of FIG. 1 or due to thermal expansion effects. If this should occur then the effect will be to displace both beams, 50 and 52, (of FIGS. 5a and 5b) equally thereby holding the distance “d” fixed. This, of course, allows the sensor readings to reject these effects and be primarily responsive to torque alone.

A Fabry-Perot etalon is an interferometer instrument having two, parallel plane reflecting surfaces for optically measuring the distance between the distance between the two surfaces. Traditionally, one of the reflecting surface is a substantially fully reflecting mirror whereas the other surface is a partially reflecting dichroic. A collimated light ray is directed through said dichroic against said mirror. FIG. 6 illustrates one method for installing a Fabry Perot etalon between the two beams, 50 and 52, to measure the differential gap d between them. Basically, a reflecting surface, 56, is secured to the top of the bottom beam, 52, and a partially reflecting dichroic surface, 58, is installed on beam, 50, facing surface 56. Both surfaces are normally traversed by a common axis. Beam, 50, also has a transverse aperture 59 to allow the passage of a collimated light ray from an optical fiber 60 mounted above it. The fiber is mounted into a secure beam 54 with a ball lens 62 between it and the first reflecting surface 56 The purpose of the ball lens is to collimate the light ray from the optical fiber before reaching the Fabry Perot dichroic lens.

This structure offers many advantages. First, the optical fiber can be mounted and maintained essentially straight with the central axis of the sensor. This eliminates any bending issues of the fiber. Secondly, the Fabry-Perot interferometer measures displacement on a non-contact basis without any resulting resistance back to the sensor. This feature provides the ability to size the mechanical components to measure very low torque levels. Next, the combination of the pressure isolation torque tube 10 and the torque to force assembly has the ability to reject vibration and temperature effects. Lastly, the complete sensor makes it possible to accurately measure small values of torque created within a pressurized well fluid environment. These environments generally comprise corrosive well bore fluids which may be pressured from atmospheric to very high values, such as 20,000 pounds per square inch. If a well is under extreme high pressure, then this may cause a significant axial strain of the pressure isolator tube and this will also be rejected by the differential output arrangement.

It should be apparent that other types of optical sensors may be substituted for the Fabry-Perot interferometer. These, however, will generally require a stress stimulus and therefore present a force load on the torque-to-force converter. These devices also generally will require either a dark termination or an output optical fiber. Meeting either of these requirements will significantly complicate their application. It should also be apparent that it is possible to introduce a second Fabry-Perot etalon into the optical fiber 60 of FIG. 6, for the purpose of measuring the sensor ambient temperature. This would allow temperature compensation of the torque sensor.

FIG. 7 illustrates a special case of this sensor. Basically, the isolation torque tube 10 of FIG. 1 has been replaced by a Bourdon tube 72. This is a special Bourdon tube in that it is configured to be symmetrical about its central axis 74 and therefore has a pure torque output response 75. This torque response 75 is coupled to a torque-to-displacement converter 40 constructed exactly as previously described with respect to FIG. 4. This will result in a differential displacement gap d modulation for the Fabry-Perot etalon and an optical response via optical fiber 60. This sensor configuration, then will provide the ability to optically measure an input pressure P with high accuracy and good rejection of temperature and vibration effects.

FIG. 8 a illustrates another fiber optic embodiment of this invention for a Fabry-Perot gap “d” and in FIG. 8b for a Bragg grating type fiber optic embodiment. Sensing element 80 in both figures are fused silica capillary tubes, which are commercially available with a precision bore to match the outer diameter of standard optical fibers 82 and 83 (FIG. 8a) or 86 (FIG. 8b). A standard fiber diameter is 125 uM (micro-meters) and a capillary is available with a bore of 130 uM. This gives the fiber a 5 uM (approximately 200 micro inches) clearance within the tube.

FIGS. 8 a and 8b also show that the fibers are bonded at each end 88 and 89 of the capillary tube 80. This bonding can be a thermal fusion as performed by a laser. If a capillary tube with 650 uM outside diameter is used then the end result of FIGS. 8a and 8b will be a miniature fused silica capsule which will perfectly match the fused silica fiber material and which is large in diameter with respect to that of the fiber. The capillary bore is small enough to prevent buckling of the fiber for compressive loads and the outside diameter is large enough to be stable against buckling for the lengths required for both of these approaches. These lengths would typically be about 7.5 mm for the Fabry Perot sensor of FIGS. 8a and 15 mm for the Bragg grating sensor of FIG. 8b. The Fabry Perot gap “d” shown in FIG. 8a is formed when the independent fibers 82 and 83 are axially aligned and bonded into place with a gap “d” end separation. However, Bragg grating 87 is written into the fiber 86 of FIG. 8b prior to being bonded into place.

FIG. 9 illustrates a basic principle common to all of the presently described embodiments of the torque sensor. This principle is based on installing a sensing element 80 between beams 50 and 52 and which spans the gap between them. The illustration of FIG. 9 is for the fiber optic load cells of FIGS. 8a and 8b. This is shown by FIG. 9 as item 80 within the active sensing span 100 between beams 50 and 52. The active area of each capillary load cell 80 is the area between the end bonds 88 of FIGS. 8a and 8b. Each end 81 of the capillary load cell can be extended, as required, beyond the end bond to facilitate mounting.

The effects of temperature expansion can be largely avoided by constructing the beams 50 and 52 of a material with a suitable temperature coefficient of expansion, such as invar. In the event that Bragg gratings are used then a second Bragg grating can be written into the fiber at point 90. This Bragg grating will not be stressed and therefore its response can be used to measure for temperature correction.

An important issue with respect to the application of Bragg gratings is that they generally do not work well in tension due to micro cracking within the fiber. This effect can be overcome by simply configuring the sensor to always operate in the compression mode, either by limiting the torque to always be on one side of zero or by installing the capillary load cell 23 with a compression bias.

A major advantage of the capillary type load cells 80 is that they convert the force between the beams 50 and 52 to a micro-displacement which is suitable for application with this torque tube/force converter approach. They also act as matching devices to provide a perfectly elastic element to couple to the micro-displacement characteristics of the optical fibers.

The embodiments of FIGS. 4-7 and FIGS. 8-10 use the Fabry Perot and Bragg grating principles, respectively, as sensors. Another invention embodiment represented by FIG. 10 comprises replacing the fiber optic element 80 of FIG. 9 with a resonating beam 92. This can be a Piezoelectric or mechanical type, such as shown in FIG. 10 and as described in U.S. Pat. No. 4,372,173. The resonator 92 of FIG. 10 comprises two beams 94 and 96 joined at their ends. The beams are caused to vibrate by piezoelectric means and their frequency will be a function of an end applied force, F. Such a resonator 92 can also be constructed with a single beam and both types are similar in operation to a vibrating string except that the beam construction will work for compressive as well as tensile forces.

In all of these invention embodiments, the sensing elements installed between the beams exhibit a very small deflection for an applied force. This characteristic allows the overall response of the torque sensor to be primarily a torque force measurement device as opposed to a torque displacement measurement type and this conforms to the original stated objective of this invention.

An additional invention embodiment is shown by FIG. 11 and illustrates a special case of this sensor. Basically, in a manner similar to U.S. Pat. No. 5,207,767, the isolation torque tube 10 of FIG. 1 has been replaced by a Bourdon tube 72. This is a special Bourdon tube in that it is configured to be symmetrical about its central axis 74 and, therefore, has a pure torque output response. This torque response 75 is coupled to a torque-to-displacement converter 40 constructed exactly as previously described. Hence, a force sensing element 80 is attached between beams 50 and 52 in the same way as outlined for FIG. 7. This structure results in a direct conversion of pressure to torque with relatively low stresses within the Bourdon tube because of the low torque displacement sensitivity of the torque sensor assembly 40.

Although the invention disclosed herein has been described in terms of specified and presently preferred embodiments which are set forth in detail, it should be understood that this is by illustration only and that the invention is not necessarily limited thereto. Alternative embodiments and operating techniques will become apparent to those of ordinary skill in the art in view of the present disclosure. Accordingly, modifications of the invention are contemplated which may be made without departing from the spirit of the claimed invention.

Claims

1. A torque measuring instrument comprising: means for converting a torsional strain about an axis to a spatial displacement across a sensor element substantially parallel with said axis;means for determining a value of said displacement; andmeans for relating the value of said displacement to a value of said strain.

2. A torque measuring instrument as described by claim 1 wherein a value of said special displacement is determined by optical means.

3. A torque measuring instrument as described by claim 2 wherein said means for determining a value of said displacement comprises a Fabry-Perot etalon.

4. A torque measuring instrument as described by claim 1 wherein said means for determining a value of said displacement comprises a capillary tube load cell.

5. A torque measuring instrument as described by claim 4 wherein optical fiber is fused to opposite ends of said capillary tube.

6. A torque measuring instrument as described by claim 1 wherein said means for determining the value of said displacement comprises a pair of light reflective surfaces, one is which is a partially reflective dichroic.

7. A torque measuring instrument as described by claim 1 wherein said means for converting torsional strain comprises a Bourdon tube spiraled about said axis.

8. A torque measuring instrument as described by claim 7 wherein a fluid pressure within said Bourdon tube is converted to said torsional strain, said torsional strain being corresponded to a value of said fluid pressure.

9. A torque measuring instrument as described by claim 1 wherein said means for converting torsional strain comprises a first element that is rotatively displaced in a first fluid environment by torsional stress imposed in a second fluid environment, said first element displacement being relative to a stationary second element positioned within said first fluid environment; a plurality of hat-section links, each having a crown panel supported by a pair of substantially equal length leg panels, a distal end of one leg panel respective to each pair being secured to said first element and a distal end of the other leg panel respective to said pair being secured to said second element; at least a pair of beam elements diametrically traversing said first element through said axis, each of said beam elements supported at substantially opposite distal ends by the crown panels of respective pairs of hat-section links.

10. A torque measuring instrument as described by claim 9 wherein a first of said beam elements overlies a second of said pair proximate of said axis with a spatial separation distance there between.

11. A torque measuring instrument as described by claim 10 wherein means for determining a value of said displacement comprises means for determining a change in said separation distance between said beams.

12. A torque measuring instrument as described by claim 11 comprising optical means for determining a change in said spatial separation distance between said beams.

13. A torque measuring instrument as described by claim 12 wherein said optical means comprises a substantially reflecting first surface being disposed on said first beam and a dichroic second surface disposed on said second beam, both of said surfaces being normally traversed by said axis.

14. A torque measuring instrument as described by claim 13 wherein a collimated fiber optic light ray is directed along said axis, through said second surface and reflected from said first surface.

15. A torque measuring instrument as described by claim 12 wherein said means for determining a change in said separation distance between said beams comprises a capillary tube load cell disposed between said beams.

16. A torque measuring instrument as described by claim 15 wherein optical fiber is fused to opposite ends of said capillary tube.

17. A torque measuring instrument as described by claim 16 wherein independent fibers are axially aligned and fused to said capillary tube ends with an end separation distance between respective fiber ends.

18. A torque measuring instrument as described by claim 1 wherein said means for converting torsional strain comprises a Bourdon tube spiraled about said axis.

19. A torque measuring instrument as described by claim 18 wherein a fluid pressure within said Bourdon tube is converted to said torsional strain, said torsional strain being corresponded to a value of said fluid pressure.

20. A torque measuring instrument comprising: means for converting a torsional stress about an axis to a lineal stress across a sensor element substantially parallel with said axis;means for determining a value of said lineal stress; andmeans for relating the value of said lineal stress to a value of said torsional stress.

21. A torque measuring instrument as described by claim 20 wherein said means for determining a value of said lineal stress comprises a Bragg grating optical fiber.

22. A torque measuring instrument as described by claim 20 wherein said means for determining a value of said lineal stress comprises a resonating beam.

23. A torque measuring instrument as described by claim 20 wherein said means for determining a value of said lineal stress comprises a capillary tube load cell.

24. A torque measuring instrument as described by claim 23 wherein independent fibers are axially aligned and fused to said capillary tube ends with an end separation distance between respective fiber ends.

25. A torque measuring instrument as described by claim 23 wherein a Bragg grating is written into a unitary fiber between said opposite capillary tube ends.

26. A torque measuring instrument as described by claim 20 wherein said means for converting torsional stress comprises a Bourdon tube spiraled about said axis.

27. A torque measuring instrument as described by claim 23 wherein a fluid pressure within said Bourdon tube is converted to said torsional stress, said torsional stress being corresponded to a value of said fluid pressure.

28. A torque measuring instrument as described by claim 20 wherein said means for converting torsional stress comprises a first element that is rotatively displaced in a first fluid environment by torsional stress imposed in a second fluid environment, said first element displacement being relative to a stationary second element positioned within said first fluid environment; a plurality of hat-section links, each having a crown panel supported by a pair of substantially equal length leg panels, a distal end of one leg panel respective to each pair being secured to said first element and a distal end of the other leg panel respective to said pair being secured to said second element; at least a pair of beam elements diametrically traversing said first element through said axis, each of said beam elements supported at substantially opposite distal ends by the crown panels of respective pairs of hat-section links.

29. A torque measuring instrument as described by claim 28 wherein a capillary tube load cell is disposed between said beams.

30. A torque measuring instrument as described by claim 29 wherein optical fiber is fused to opposite ends of said capillary tube.

31. A torque measuring instrument as described by claim 30 wherein a Bragg grating is written into a unitary fiber between said opposite capillary tube ends.

32. A method of measuring torque comprising the steps of: converting a torsional strain about an axis to a spatial displacement across a sensor element substantially parallel with said axis;determining a value of said displacement, and,relating the value of said displacement to a value of said strain.

33. A method of measuring torque as described by claim 32 comprising the step of optically determining the value of said displacement.

34. A method of measuring torque as described by claim 33 wherein a Fabry-Perot etalon optically determines the value of said displacement.

35. A method of measuring torque as described by claim 33 wherein a Bragg grating written onto an optical fiber is used to optically determine the value of said displacement.

36. A method of measuring torque as described by claim 32 comprising the step of determining the value of said displacement by a evaluating a frequency change in a resonating beam that is distorted by said spatial displacement.

37. A method of measuring torque as described by claim 32 wherein said spatial displacement is transferred to a capillary tube load cell for determination of said displacement value.

38. A method of measuring torque as described by claim 37 wherein said torsional strain is induced by fluid pressure into a Bourdon tube wound about said axis.