System and Method for Determining Airfoil Characteristics From Coordinate Measuring Machine Probe Center Data

The present invention relates generally to coordinate measuring machines, and more particularly to using coordinate measuring machine probe center data to determine airfoil characteristics.

Airfoils are used typically in aircraft engine blades and vanes, and generally have highly sculptured complex surface geometries which must be manufactured to meet precise engineering characteristics. The tolerances on the characteristics are tightly specified in order to meet performance requirements. The outside shape of the airfoil can be determined by manually inspecting the airfoil with a hard gauge such as a guillotine gauge. However, the hard gauge is very subjective and the results vary from one user to another. Another approach in determining the outside shape of the airfoil is to analyze probe center points taken on a section-by-section basis from a coordinate measuring machine (CMM). The CMM approach is computationally intensive and is highly dependent upon accurate, ordered and well distributed probe touch points. Typically, manual methods such as graphically or textually deleting the offending points from the electronic file are used to eliminate errant CMM data and to order the points in a consistent orientation about the airfoil on a section-by-section basis. Planar touch point correction of the CMM probe touch points and subsequent airfoil characteristics are determined by using continuous geometry and curve-fitting approximation methods that make assumptions about the distribution of the CMM probe center data, nominal and actual airfoil shapes. As mentioned above, this method is computationally intensive and is highly dependent upon accurate, ordered and well distributed probe touch points, which is not very practical for a typical CMM.

Therefore, it is a primary objective of the present invention to provide a system and method for using CMM probe center data in conjunction with a nominal geometric model of the inspected airfoil part to accurately derive the engineering characteristics of the inspected part without using predetermined assumptions.

Another object of the present invention is to provide a system and method for deriving characteristics that are not computationally intensive and that account for the practical realities associated with CMM measurements such as out-of-order probe center points and probe contact at unexpected locations.

Still another object of the present invention is to accurately derive engineering characteristics of an inspected part to improve quality assurance with parts.

The present invention attains these objects by using unordered probe center data points taken from a CMM and ordering the points in a consistent direction about the airfoil on a section-by-section basis. The ordered probe center points are then corrected for surface touch point inaccuracies. Next, the surface touch points are corrected for planar inaccuracies. Discrete inset points are derived from the planar touch points, which are used to determine the airfoil shape characteristics. This approach is insensitive to errant probe center data about the body of the airfoil which may occur from premature contact between the CMM probe and the part or fixture. Furthermore, any sections which cannot be analyzed due to an insufficient amount of data are detected but do not affect the analysis of any other airfoil section.

Thus, in accordance with the present invention, there is provided a system for determining characteristics of a part having a complex surface geometry. A fixture holds the part while a coordinate measurement machine having a ball probe measures external coordinates along the part. The external coordinates are a set of unordered probe center points. A computer is operably connected to the coordinate measurement machine and has a model of a nominal part stored in memory. The computer includes an ordering means for ordering the probe center points about the part in a plurality of sections. An aligning means aligns the ordered probe center points for each section to a corresponding section of the nominal part. The aligned probe center points are then planar corrected to derive planar touch points and the characteristics of the part are derived from the planar touch points.

While the present invention will hereinafter be described in connection with a preferred embodiment and method of use, it will be understood that it is not intended to limit the invention to this embodiment. Instead, it is intended to cover all alternatives, modifications and equivalents as may be included within the spirit and scope of the present invention as defined by the appended claims.

FIG. 1 is a representational view of the system used in the present invention;

FIG. 2 is a flow-chart describing the method in the present invention to determine part characteristics;

FIG. 3 is a perspective view of an airfoil;

FIG. 4 shows a two-dimensional sectional view of the airfoil;

FIG. 5 shows a two-dimensional sectional view of the airfoil with triangulations;

FIG. 6 shows a two-dimensional sectional view of the airfoil containing a set of ordered probe center points;

FIG. 7 shows a close up view of the airfoil with a CMM ball probe;

FIG. 8 is a flow-chart describing the touch point correction used in the present invention;

FIG. 9 shows a two-dimensional sectional view of the ordered probe center points and the nominal section geometry;

FIG. 10 shows a two-dimensional sectional view of the ordered probe center points aligned with the nominal section geometry;

FIG. 11 shows a two-dimensional sectional view of the ordered probe center points aligned with the nominal section geometry and estimated surface normals;

FIG. 12 shows a two-dimensional sectional view of the airfoil with planar touch points; and

FIG. 13 shows a two-dimensional sectional view of the airfoil with a meanline derived from inset points.

FIG. 1 is a representational view of the system 10 used in the present invention. A part 12 is held in place by a fixture 14. For purposes of illustration, the part described in the present invention is an airfoil, but could be any part with a planar surface, a non-planar surface, or any shaped part having a complex or non-complex surface geometry. A programmable CMM 16 having a ball touch probe 18 measures external coordinates along the part 12 in the x-y-z direction. The ball touch probe is a well known device. A computer 22, preferably, a general purpose computer is operably connected to the CMM and programmed to control the movement of the ball touch probe 18, record measurements taken from the ball touch probe, and determine the part characteristics as described below in further detail. In addition, the computer contains a memory 24 that contains a geometric model of a nominal part 26 corresponding to the part 12.

In operation, the computer 22 moves the CMM 16 in three orthogonal directions x, y, and z about the part 12. The ball touch probe 18 generates a set of unordered probe center points for the part at a given coordinate. The unordered set of probe center points are sent to the computer for processing, which is described in further detail. The processed data is used to derive the engineering characteristics of the part in a manner described below in further detail.

The engineering characteristics are determined in accordance with the flow chart set forth in FIG. 2. In addition, FIGS. 3-13 provide graphical representations illustrating the results occurring at various steps in the flow chart of FIG. 2. The computer receives a set of probe center points at 30. FIG. 3 shows a perspective view of the airfoil part 12 with a plurality of probe center points 64. The probe center points are unordered, but are identified by nominal sections of interest 66. These points are assumed to be in the same reference frame as the nominal part 26 stored in the memory 24 (i.e., a datumed part). Along with the probe center points, the probe radius is recorded.

Next, the probe center points are sorted by nominal section planes at 32. FIG. 4 shows a two-dimensional view of a section 66 from the airfoil 12 with a plurality of probe center points from the ball touch probe 18 thereat. Each section 66 of the airfoil contains a plurality of probe center points and are processed in a section-by-section manner. At 34, the probe center points at each section are triangulated by using a triangulation method such as a Delaunay triangulation or the like. In the triangulation, the probe center points 64 are connected to form triangles 68 as shown in FIG. 5. Note that the triangulated section in FIG. 5 contains vertices 75 that are interior and exterior to the section 66, respectively. Interior and exterior vertices are considered to be errant CMM probe hits and are excluded at 36. Once all of the errant points have been excluded, the boundaries of the triangles are traversed in a counter-clockwise fashion at 38, removing segments 72 connecting the vertices 70. Removing the segments from the vertices 70 results in a set of ordered probe center points 73 as shown in FIG. 6.

In order to account for inaccuracies associated with the ball touch probe 18 contacting the airfoil 12, the ordered probe center points are corrected and transformed into planar touch points on the actual airfoil surface at 40. FIG. 7 illustrates the reason why planar touch points are necessary. In particular, the ball touch probe 18 is shown contacting the airfoil at an actual touch point 74. However, only the probe center point 76 is received by the computer and not the actual touch point or the desired touch point 78. Thus, it is necessary to determine the actual touch point and the desired touch point for all of the ordered probe center points.

Touch point correction is performed using the ordered probe center points along with the nominal geometry for each section and is described in the flow-chart set forth in FIG. 8. At 44, the bounds of the nominal section geometry and the ordered probe center points are determined by finding the minimum and maximum x and y coordinate positions of all of the points. FIG. 9 shows a two-dimensional sectional view of the ordered probe center points 73 and the nominal section geometry 26. A discrete offset is performed at 46 on the nominal section data using an offset value equal to the probe radius. The offset is used to compensate for CMM measurements on part 12.

Next, a coordinate transformation matrix is determined at 48 and used to align (i.e., translation, orientation and scaling) the ordered probe center points to the nominal section data at 50. The coordinate transformation is preferably a 4.times.4 matrix which is well known to skilled artisans. FIG. 10 shows a two-dimensional sectional view of the ordered probe center points 73 aligned with the nominal section geometry 26 after the coordinate transformation has been applied. Discrete offsets are then determined at 52 for the transformed probe center points and the nominal section data. In particular, a sufficiently large offset value is applied to the transformed probe center points and the nominal section points. This offset value must be small enough to avoid having the resulting shape collapse upon itself. The offset value used for the transformed probe center points is smaller than the offset value applied to the nominal section points by the probe radius. These offsets are used to improve the correlation between the nominal and inspection points.

Each transformed probe center point is then matched at 54 to the nominal section data to estimate a surface normal 80 by a weighted average of the surface normals of the two closest nominal points (see FIG. 11). The original probe center points are then translated at 56 by one probe radius in the direction opposite to the estimated surface normals 80, to produce equivalent touch points. To correct the touch points to the desired touch points 78 or the nominal plane of interest, a tangent plane to the nominal surface at the touch point is determined at 58. After computing the tangent plane, a cross-spline vector in the tangent plane is computed. A cross-spline vector is a vector perpendicular to both the normal vector and the tangent vector at the point of interest. The cross-spline vector is used to approximate where the surface is, both above and below the section of interest. The intersection between the cross-spline vector and the nominal plane of interest is used to translate the touch points to the nominal plane of interest. FIG. 12 shows a two-dimensional sectional view of the airfoil 12 with the computed planar touch points 78.

After the planar touch points have been determined, the engineering characteristics of the airfoil are determined at 60. The engineering characteristics are derived by performing a rough characterization of the airfoil length and width. Then using a starting value for a discrete inset operation based on the rough characterization, the planar touch points are successively insetted with large values until an inset is no longer possible, yielding a maximum thickness M. FIG. 13 shows a two-dimensional sectional view of the derived section geometry having a maximum thickness M formed from inset points 82. In particular, at each inset operation, an inset curve will collapse on itself at opposing ends, in effect, pinching off the ends of the inset curve. These opposing points are points on a meanline 88. The discrete inset process continues from the original starting value using successively smaller inset values until the inset planar touch points no longer collapse to points at either ends 84 or 86 of the airfoil 12. All of the inset points 82 from each inset operation are joined to form an airfoil meanline 88. The intersection between the meanline and the connected planar touch points result in leading and trailing edge points. Airfoil characteristics such as leading edge thickness, trailing edge thickness, basic chord length, etc., are determined from the meanline, leading and trailing edge points, and the planar touch points using well known geometric constructions such as computing distances along the meanline, projecting perpendicular vectors about each side of the meanline, performing line-line intersection, and computing the distance between two points. After the characteristics have been determined, they are outputted at 62.

It is therefore apparent that there has been provided in accordance with the present invention, a system and method for determining airfoil characteristics from coordinate measuring machine probe center data that fully satisfy the aims and advantages and objectives hereinbefore set forth. The invention has been described with reference to several embodiments, however, it will be appreciated that variations and modifications can be effected by a person of ordinary skill in the art without departing from the scope of the invention.