Force transducer flexure with conductors on surfaces in the neutral bending plane

(24) Running day

PATENT AUTHORITY (62) Stamansokan number

02-07-08

82-07-08 (86) International filing day (86) Filing date for European patent application (30)

Application received as:

□ Swedish patent application

O completed international patent application with number

Q converted European patent application with number

81-07-14 US 283340

<td> (71)</td><td>Applicant</td><td>Sundstrand Data Control Inc, Redmond Wash US</td>

<td> ( 72 )</td><td>You ppfi η nasty</td><td>RA Hanson, Woodinville Wash</td>

<td> ( 74 )</td><td>Agent</td><td>Holmqvist L</td>

<td> ( 54 )</td><td>Name</td><td>Flexible body for a power transducer with conductors located on the surface of the flexible body in its neutral bending plane</td>

<td> (56)</td><td>Quoted publ</td><td>icons: —-</td>

<td> (57)</td><td>Summary:</td><td></td>

Flexible member of a power transducer, such as an inertial control accelerometer for attaching a force sensing element (30) to a mounting base (32). The flexible member (12) comprises one or more flexible portions (60, 62) having one or more excavated surfaces (72, 74) which are substantially substantially coincident with the neutral flexural plane of the flexible member. Electrically conductive coatings (47) on the excavated surfaces provide electrical connection to components (42, 43, 45) located on the force sensing element. The conductive coatings (47) on or near the neutral bending plane of the flexible portions minimize bending moments caused by stresses formed between the conductive coatings and the flexible member, which in turn can lead to biasing errors. At the same time achieve

DB 647289 design of the flexible member the desired strength and resilience

<img file="SE451897B_D0001.tif" />

<img file="SE451897B_D0002.tif" />

The numbers mom parentheses indicate international identification code, INID code Letters in clamps indicate international document code

451 897

The present invention relates to the field of force sensitive sensors, such as inertia control accelerometers. More particularly, the invention relates to a flexible means for connecting a force sensing element to a mounting base for pivotal or translational motion.

In this type of power transducer, such as the accelerometers described in U.S. Pat. Nos. 3 702 073, 4,182,187 and 4,250,757, a force sensing element is attached to a mounting base or ring by a flexible member which allows the element to move in dependence on small forces relative to the base. In such an accelerometer, the flexible member may have a dual or bifilar structure consisting of a pair of thin planar elements.

To provide electrical connections to components located on the sensing element, flexible conductors between the base and the sensing element may be used, or a thin film of conductive material may be deposited or coated directly on the flexible member or on a non-conductive coating on the flexible member. the member of the flexible member itself is electrically conductive. When such materials are coated on the flexible member, stresses build up in the flexible member due to the various temperature coefficients of the flexible member and the conductive materials, or the coating process itself. These voltages, in turn, result in forces which aim to deflect the force sensing element from a neutral position. In a servo loop encoder that supplies a restoring force to retain the sensing element in the neutral position, biasing errors are generated as a result of these voltages. In an open loop encoder where the deflection of the sensing element is measured, biasing errors are also generated.

In the sensors using conductive coatings, attempts have been made to quench the film stresses by applying films to the same extent on the upper and lower surfaces of the flexible portions. This construction reduces the errors to some extent, but it requires careful balancing during the coating process so that the film thickness is equal to both.

451 897 sides of the flexible section. Furthermore, this balancing is dependent on the stability of the film voltage with respect to time and is also dependent on other factors such as the ambient temperature, the purity of the material and surface impurities.

In general, prior art sensors have found it desirable to use the thinnest possible flexible means, taking into account the strength and requirements of elasticity for proper functioning, so that the voltage effects leading to bias errors are minimized. However, it has been found that the spring capacity, either the perpendicular or linear, provided by a flexible member is proportional to the cube of the thickness t of the flexible member, while the bending moment of the flexible member due to the stresses caused by the coating of conductive strips is proportional to t. For example, if the thickness of the flexible member is reduced by 30 °, so that the angular spring capacity is changed from 20 ° / radius to 7 ° / radius, the failure moment due to the stress effects in the conductive coating is reduced by a factor of only 1.42. Thus, it will be appreciated that the lower limit of acceptable resilience of a conventional flexible member is reached long before the error torque has been reduced to an acceptable value. Thus, for these types of flexible members, compromises must be made between obtaining the desired resilience and strength of the flexible member and minimizing the stress effects leading to failure.

According to the present invention, a flexible member is provided with an electrically conductive coating on surfaces located substantially on or in the neutral bending plane of the flexible member.

Since the conductive material is located on a surface which is mainly in the neutral bending plane of the flexible member, stresses in the interface between the conductive film and the flexible member do not cause biasing errors. This is because the bending moment caused by such stresses is proportional to the distance from the interface to the neutral bending plane. Since this distance is essentially zero, the bending moment is minimized and the bias errors are significantly reduced, if not complete.

The flexible member may consist of one or more portions, each containing one or more channels, each having a recessed surface located mainly on the neutral bending plane of the flexible member.

Since the conductive film is always located on or adjacent to the neutral bending plane, the magnitude of the error supplied by the conductive film is essentially independent of the thickness and the total length and width of the flexible member. Thus, the flexible member can be designed to provide sufficient strength and desired resilience to the accelerometer without causing significant biasing errors due to film stresses.

451 897

Similar solutions to this problem are set forth in the Swedish patent applications 8204220-1 and 8204219-3 published in parallel herewith.

The invention is explained in more detail below with reference to the drawings which show some preferred embodiments of the invention. Fig. 1 is an exploded perspective view and partially in cross-section of an accelerometer incorporating the flexible member of the present invention. FIG. 2 is an enlarged fragmentary plan view of the flexible member of FIG. 1. FIG. 3 is a fragmentary perspective view taken substantially along line 3-3 of FIG. 1. Fig. Is a cross-section through a flexible member having different width channels. Figs. 6 are fragmentary perspective views similar to Fig. 3, showing further embodiments of a bifilar flexible member. Figures 7 and 8 are fragmentary perspective views similar to Fig. 3 showing two embodiments of a single member flexible member.

Figures 1 and 2 show a power transducer in the form of a servo-controlled accelerometer 10, which comprises a flexible member 12 according to the present invention. The accelerometer 10 is of the type disclosed in U.S. Patent No. 3,702,073, although the invention may be used in other force sensing sensors using angular motion or linear motion of a force sensing element.

The accelerometer 10 consists of a pair of cylindrical body elements 14a, 14b and an inertial mass assembly 16 arranged therebetween.

The body elements 14a, 14b are essentially identical and therefore only the body element 14a is described. The body member 14a comprises a cylindrical body wall 17 having an inwardly extending edge 18, which forms a magnetic pole piece and extends around the bottom and forms a bottom 19. The pole piece 18 has a cylindrical inner wall 20 which forms a recess 22. A cylindrical permanent magnet 24 is attached to plate 19 in recess 22. The permanent magnet 24 has an outer peripheral surface spaced from the inner cylindrical wall 20 to form an annular space 26 therebetween.

The inertia mass assembly 16 comprises a force-sensitive element or tongue 30 which is hingedly mounted by the flexible members 12 at a mounting base or ring 32.

In the accelerometer shown in the figures, the flexible member 12 allows the tongue to move arcuate relative to the mounting ring 32. However, it will be appreciated that flexible members 12 can be used in a sensor using linear motion of the sensing element along the axis of the sensor 10.

At the upper and lower surfaces 40, 41 of the tongue 30 are a pair of force-restoring coils or torque coils 42, 43. The torque coils 42, 43 are wound on bobbins which fit into the annular space 26 formed in each body member 14a, 14b. , when the various parts of the acceleroΐΐΐ

451 The 897 meters 10 are assembled.

On the upper surface 40 of the tongue 30, a layer of conductive material 45 is provided. A similar layer of conductive material is deposited on the lower surface 41 of the tongue 30. These electrically conductive layers form a pair of capacitor plates which cooperate with a surface 21 of the pole piece 18 and a corresponding surface of the pole piece of the element 14b, in a manner described more detailed below.

Three mounting pads 34 (one of them not shown in the figures) are located on an upper surface 36 of mounting ring 32. Three additional mounting pads are located axially opposite mounting pads 34 on a lower surface 38 of ring 32.

The mounting ring 32 is secured between the body elements 14a, 14b such that an edge of the wall 17 of the cylindrical body and a corresponding edge of the body element 14b abuts the mounting pads, the torque coils 42, 43 being received in the annular space 26 and a corresponding annular space in the body element 14b.

A pair of variable capacitors 48, 49 are formed in accelerometer 10, one capacitor consisting of surface 21 and coating on lower surface 38 and the other consisting of a surface corresponding to surface 21 on the pole piece of element 14b and coating 45 on upper surface 36 of tongue 30. 8 *

The conductive bearings on the upper surface 40 and the lower surface 41 as well as the torque coils 42, 43 are coupled to outer circuits by four conductive strips 47a-d extending to the ring 32 over the flexible member 12. The electrical connections to the outer circuits occurs from the ring 32 via four contact pins (not shown) located in the body wall of the body elements:

14a, 14b.

When accelerometer 10 is subjected to acceleration along its axis, tongue 30 moves relative to ring 32 and body elements 14a, 14b, which in turn causes a change in capacitance of capacitors 48, 49. The change in capacitance is sensed by a servo circuit (not shown). which in turn couples a signal proportional to the change in capacitance to the torque coils 42, 43. The resulting magnetic field of the torque coils 42, 43 cooperates with the magnetic field of the permanent magnets of the body elements 14a, 14b to counteract the displacement of the tongue 30. The current required by the torque coils 42, 43 to maintain the tongue 30 in a neutral position corresponds to the accelerometer force. exposed to.

For further description of the accelerometer 10, reference is made to U.S. Patent 3,702,073.

The preferred embodiment of the flexible member 12 of the present invention is shown in Figure 3.

451 897

The flexible member has a double-sided or bifilar, freestanding, hinge-like configuration and consists of a pair of flexible portions 60, 62 extending between the tongue 30 and the mounting ring 32. The portions 60 and 62 are separated by an intermediate opening which also extends between tongue 30 and mounting ring 32.

The flexible portions 60 and 62 have substantially the same cross-section and therefore only the portion 60 is described.

The portion 60 includes an upper and a lower surface 64 and 66 which are substantially parallel to one another, and a pair of edges 68, 70 which are generally perpendicular to the surfaces 64, 66. The surfaces 64, 66 may also be non-parallel and still the channels on the neutral axis can be used.

The flexible portion 60 has a pair of channels 71, 73 extending into the portion 60 from the surfaces 64, 66. The channels 71, 73 have grooved surfaces 72, 74 which are substantially parallel and coincide with the neutral bending plane of the flexible portion. 60th

The neutral bending plane is defined as the plane which is not subjected to stretching or compression when the flexible member bends. For a flexible member having a uniform rectangular cross-section consisting of two parallel surfaces and two edges as shown in Fig. 3, the neutral bending plane consists of points which are equally spaced from both surfaces of the flexible member, ie the plane located in the middle. between the surfaces of the flexible member.

Since the channels do not have the same width on each individually flexible portion, the neutral bending axis (plane) will not divide the flexible member into two equal parts, but will instead be displaced towards the narrower channel. The cross-section of such a flexible member is shown in Fig. 4. In such a case, the position of the neutral axis will be determined by the formula:

Y = Yo [_- (Δρ / Aul) + [(Δρ / Aul)² - 1]] (1) where Y is a coordinate of the center of the flexible body at the bottom of each channel where Δρ / Δυΐ&gt;. 1. Yo is | of the total thickness of the flexible member.

is 2x (the width of the flexible body) - (the sum of the channels width). &amp; ul is (the width of the upper channel) - (the width of the lower channel).

It should be noted that the current neutral bending plane is somewhat different from parallelism with the plane of the flexible member due to the asymmetry around the center line of the individual flexible member, but since this effect is normally small, each channel may be located so that the neutral bending plane passes through the center of the the channel if the angular rotation were to be

451 897 considerable.

The ducts 71, 73 also have sidewalls 71a, 71b and 73a, 73b. Although the channels shown in Figs. 1-3 have flat walls 71a, 71b and 73a, 73b which are perpendicular to surfaces 64, 66, these surfaces may alternatively be chamfered or have a curved shape with a small radius. Further, the flexible portions may be chamfered or curved in the region of engagement with the tongue 30 and the mounting ring 32 to form a smooth transition therebetween unlike the sudden transition of Figures 1 and 2. In the preferred embodiment, the tongue 30, mounting base 32 and flexible member 12 are formed in a continuous piece of molten quartz etched or otherwise treated to the desired shape. For a detailed description of the manufacturing process for the flexible member 12, reference is made to the parallel Swedish application no. 8204221-9.

Once the excavated surfaces 72, 74 have been fabricated, a thin layer of conductive material, such as gold, is applied along the length of the excavated surfaces of both flexible portions to extend between the tongue 30 and the mounting ring 32.

Alternatively, the flexible member 12 may be made of an electrically conductive material, in which case an electrically non-conductive coating is deposited on the flexible member 12 before the conductive strips 47 are applied.

Since the tensile bending moment of the film is proportional to the distance at which the conductive strips 47 are located from the neutral bending plane or axis, a force acting at the neutral bending plane parallel to its surface will not produce any bending moment, since the distance would be = zero. Thus, a conductive strip applied directly to the neutral bending shaft of the flexible member will not cause any failure of the output signal from the transducer due to voltages in the interface between the conductive strip and the flexible member. Since the conductive strip has a finite thickness, ideally, but not necessarily, the channel surfaces 72, 74 should be located past the neutral axis by half the thickness of the strip, so that the strip itself is centered around the neutral axis.

By way of example, the following dimensions have been found to give satisfactory results.

Reference numeral in Fig. 3

Description

Dimensions (in mm)

1st Total width of flexible member 12 9.14

2nd The width of each lot

3,81

3rd The distance between the edge 68 and the channel 71 is 2.29 c

<img file="SE451897B_D0003.tif" />

4th The width of the excavated surfaces 72, 74

0,89

<td></td><td> 451</td><td> 897</td>

<td>e</td><td>7 5. The thickness of portions 60, 62</td><td> 0,03</td>

<td>f</td><td>6. Depth of neutral bending plane from surfaces 64, 66</td><td> 0,015</td>

<td>g</td><td>7. Length of flexible member 12</td><td> 2,54</td>

<td>hrs</td><td>8. The distance between adjacent edges of portions 60, 62</td><td> 1,52</td>

<td></td><td>9. Thickness of the conductive strips 47</td><td> 0,00026</td>

The design of the flexible member 12 in Fig. 3 is hereinafter referred to as antisymmetric, i.e. the flexible portions have substantially the same cross-section, but are not mirror images of each other.

The thickness, length and width of the flexible portions can be selected to provide the appropriate strength and desired suspension properties for the proper operation of the accelerometer 10 without significantly increasing the bias error. Thus, unlike prior art sensors, the magnitude of the error supplied by the conductive strips 47 is substantially independent of the dimensions of the portions. It will be appreciated that any channel can be designed to include more than one insulated conductor.

Referring to Fig. 5, there is shown an alternative embodiment of the flexible member 12 which can be used if the type of the particular power generator

design requires it to increase the load resistance. This embodiment differs from that of Fig. 3 in two respects. First, a recessed surface 82 of a channel 83 is in a flexible portion 80 adjacent to an edge surface 84 without any intermediate portion of an upper surface 86. Second, portion 88 is a mirror image of portion 80, i.e., there is symmetry about it. a plane which intersects the space between the two portions 80, 88 and is perpendicular to the upper surface 86 and a lower surface 90.

The channels shown in Fig. 5 have chamfered side walls 83a and 85a and 85b. However, these walls may be flat surfaces or curved surfaces as mentioned above.

The excavated surface 82 and an excavated surface 87 are again located mainly on the neutral bending plane or axis of the portions 80, 88. Also, the conductive strips 47 are located on the excavated surfaces 82, 87 and thus do not cause stresses occurring in the interface between the flexible member and the conductive material a corresponding bending moment in the flexible member.

Referring to Figure 6, there is shown a third embodiment combining selected features of the embodiments of Figures 3 and 4. This design may be desirable where less load resistance is required but symmetry is still desirable due to the geometry of the structure.

•0

As can be seen from Fig. 6, a pair of portions 100, 102 are symmetrical, ie

451 897 mirror images similar to the embodiment of FIG. 5, but having the main configuration of the portions shown in FIG. 3. It should be noted that the portions have substantially the same dimensions as the portions shown in FIG. 3.

Fig. 7 is a further embodiment of the invention similar to the embodiment of Fig. 5 but without a separating opening between the flexible portions. This design can be used in a bifilar design where more than four electrical conductors are required and the channels cannot accommodate more than one conductor, or also in a single flexible portion design, where, for example, there is insufficient space for a bifilar design, the larger one being channel 110 can carry two insulated conductors. A single flexible portion 104 is shown with channels 106, 108 and 110 formed in excavated surfaces, 107, 109 and 111. Two of the channels 106, 108 are located in a surface 114 of the portion 104 while the remaining channel 110 is located in the other surface. 116 such that the surface 111 is located on the same plane as the surfaces 107 and 109. Thus, the surfaces 107, 109 and 111 are each substantially located on the neutral bending plane of the flexible portion and the conductive strips 47 are located on these surfaces. 0m the width of the channel 107 plus the channel 109 = the width of the channel 111, the neutral bending plane or axis is the plane dividing the flexible member 104 into two equal parts. Otherwise, the neutral bending shaft is positioned according to a formula similar to that of two channels given in Equation 1. This configuration also has no neutral shaft rotation relative to the plane of the flexible member.

Referring to Fig. 8, another embodiment of a single flexible portion with minimized asymmetry is shown. More than such a flexible portion can also be used. The portion 120 includes two pairs of channels which have aligned grooved surfaces 122, 124 and 126, 128. Two thin portions 130, 132 are formed, one between the surfaces 122, 124 and the other between the surfaces 126, 128.

The excavated surfaces 122, 128 are mainly located on the neutral plane of the flexible portion 120, while the excavated surfaces 124, 126 are slightly offset from the neutral plane. Thus, the application of conductive materials to the excavated surfaces 122, 124, 126 and 128 means that voltages occur in only two of the four channels, i.e., voltages are formed only because of the conductive materials in the excavated surfaces 124, 126. Alternatively, conductive material may be applied only to the surfaces 122, 128 which are on the neutral axis. The location of these surfaces can be determined via a formula similar to the formula for two channels according to Equation 1. In either case, the resilience and strength of the flexible member is provided mainly by the relatively thick portions 140, 142, 144 which are located on either side of the thin portions 130, 132.

451 897

As indicated above, the flexible structure of Fig. 8 can be used in a bifilar configuration, the location of which may be either symmetrical or antisymmetric. In an antisymmetric design, two identical flexible members of the type shown in Fig. 8 can be used with conductive material arranged only on the surfaces 122 and 128 located on the neutral axis of both flexible portions.

In another embodiment, the flexible members of Fig. 8 may be designed so that both of the excavated surfaces on one side of the flexible portion, such as surfaces 124 and 128, are located on the neutral plane of the flexible portion. Furthermore, flexible portions of this type need not have parallel surfaces, but they may be inclined, for example, to the force sensing element to provide a flexible member of constant strength.

It should also be noted that the principle described above for reducing the stress effect of a coating on a flexible element can be used in a flexible construction which is not necessarily the primary support element for the force sensing element. Such a construction which connects a force sensing element to a mounting base must still have a certain structural stiffness but provide a minimum of biasing forces or torque. Further, it should be noted that although the flexible structures shown in Figures 1-4 show the thin portions of the flexible portion integral with the thicker portions, it is not necessary for some applications of the invention that the thin portions 71, 73, carrying the conductive strips 47 are physically connected to the carrier portions.

The various embodiments of the invention described above indicate that the various channels aligned with the neutral bending shaft of the flexible member are located on the flexible portions which provide the principal support of the force sensing element 30 relative to the mounting base 32. However, the principle of applying an electrically conductive material, such as the conductive strips 47 along the neutral bending shaft of a flexible member, can be applied to an unloaded flexible member or bendable element as well.

It should also be noted that although the various embodiments of the invention shown include the channels extending from the force sensing element 30 to the mounting ring 32, the invention includes the placement of the electrically conductive material 47 on just a portion of the neutral bending shaft of the flexible member. preferably where the minimum strength of the flexible member is required to minimize the stresses induced by the conductive strips 47.

451 897