Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
2002-11-26
2004-12-14
Picard, Leo (Department: 2125)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C700S110000, C700S160000, C700S173000, C703S001000
Reexamination Certificate
active
06832128
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a method for rendering and/or evaluating a surface quality of a workpiece based on program data used for machining. These program data include a set of points describing the surface, wherein the set of points describes points along the path of space curves. The invention also relates to a method for optimizing the surface quality.
High-speed machining or high-speed cutting (HCS) has recently become more widespread in milling operations. The development of new technologies, such as high frequency spindles, modern cutting materials and highly dynamical, digital feed drives in numerically controlled machine tools and robots has resulted in increased use of HSC machining.
In particular, the construction of tools and molds has experienced significant advantages, such as reduced machining time, lower machining costs and shorter throughput times as compared to conventional machining of materials.
The high precision of HSC installations can also eliminate the need for manual finishing, in particular when free-form surfaces are milled.
The free-form surfaces are typically curved three-dimensional surfaces, for example fenders of an automobile or turbine blades. When these workpieces are milled, a very high measurement accuracy and surface quality is generally required.
However, the HSC method can also cause problems in certain applications. For example, in the construction of molds, the surface quality may no longer be guaranteed due to unpredictable errors. The precise, highly dynamical drives of the HSC machines appear to cause more errors than conventional drives.
The underlying cause could be determined quickly by comparing two workpieces milled on different machines. The machine that produces a poor surface quality would then be responsible for generating the machining errors. Such behavior, however, can generally not be determined. There are obviously certain factors which are influenced by different data processing or construction of these machines.
Older, less dynamic drives react rather sluggishly to the (drive) control input variables. This smoothes out small irregularities or flaws of the control input variables. As a result, rather soft, homogeneous surface structures are obtained which, however, may have greater measurement tolerances (smoothing).
However, a HSC machining process with its precise, highly dynamic drives can transfer the control variables more exactly. Smaller flaws which were previously smoothed out then become increasingly visible. Rough, inhomogeneous (hard outlined) surface structures are the result.
This processing method can achieve a higher accuracy at the expense of sometimes unacceptable surface quality.
The present invention combines the advantages of HSC machining with the high quality of conventional machining. This is accomplished by determining the underlying causes for the inadequate quality of HSC machining and by proposing improvements.
The path from a virtual model to the milled workpiece will hereafter be described into form of a process chain. This process chain is a linear sequence of individual decoupled process steps. It is therefore necessary to determine the causes for the inadequate surface quality across the entire process chain.
The process chain can be subdivided into four larger main areas.
FIG. 2
shows such a simplified process chain.
Three well-defined points (CAD model, NC data, control variables) exist within the process chain. The analysis of the NC input data (NC program) is an object of the present invention.
Except for a few special cases, the process chain depicted in
FIG. 2
with the processing paths (CAD construction→NC controller→Machine Drives) is presently almost exclusively used. Processing is typically preformed using linear sets. However, several manufacturers of control systems have since several years been able to process spline-based workpiece models without prior conversion into linear sets. In the future, the conversion could be eliminated entirely (by processing the CAD data without converting them first in the NC controller). Since an intermediate step is eliminated or at least significantly simplified, possible conversion errors are also reduced. However, until then, conventional processing methods will most likely be used to ensure compatibility.
The surface quality can be improved by intentionally intervening in the control variables. However, this intervention either prolongs machining times or replaces the old errors with new errors. Moreover, optimizing the control variables is quite costly due to the complexity of the control path. An analysis of the NC input data (workpiece data) is therefore a next step for reducing errors. Recognizing the errors is a first step in their elimination.
Until now, surface quality has been rendered by providing a surface grid of the set of points describing the surface and by then evaluating the rendition. This rendition view, however, provides only a limited view of the surface quality, because it is difficult to recognize the large number of the surface quality features of complex CNC program data.
It would therefore be desirable and advantageous to provide an improved process, which obviates prior art shortcomings and which can be used to analyze the surface quality of a workpiece, that is still to be manufactured, based on CNC program data before the workpiece is actually machined, and to thereby quickly and easily recognize possible error locations or in accuracies. In addition, an effective method for visualizing the determined surface quality should also be provided.
SUMMARY OF THE INVENTION
According to one aspect of the present invention, a method for rendering and/or evaluating the surface quality of a workpiece based on program data used for machining includes determining and rendering the associated normal vectors for a plurality of adjacent points along the machining path, wherein regions with a high surface quality are indicated by normal vectors that are oriented essentially in the same direction, whereas flaws in the resulting surface are indicated by normal vectors pointing in different directions.
Such normal vectors are preferably determined by forming two vectors from three consecutive path points and arranging the normal vector of a center path point as a vector product perpendicular on a plane spanned by the two vectors, whereby the orientation of the normal vector relative to a side of the plane is selected depending on the direction of curvature of the space curve at the corresponding path point.
According to another feature of the present invention, so-called angle-bisecting vectors can be employed instead of the normal vectors. The corresponding angle-bisecting vectors are determined and rendered for a plurality of adjacent path points. Regions with a high surface quality are indicated by angle-bisecting vectors pointing essentially in the same direction, whereas flaws in the resulting surface are indicated by angle-bisecting vectors pointing in different directions. Preferably, such angle-bisecting vectors can be determined by forming two vectors from three consecutive path points and arranging the normal vector of a center path point as a vector product perpendicular on a plane spanned by the two vectors, and by rotating the normal vector into the plane by an angle of 90°, so that the angle-bisecting vector is located at half the angle between the two vectors.
According to another feature of the present invention, all determined normal vectors of the path points may be rendered with their center located at one point. Conclusions regarding the surface quality can be drawn based on the distribution of the normal vectors, in that regions of high surface quality are indicated by the essentially overlapping normal vectors, whereas flaws of the resulting surface are indicated by a scatter of the normal vectors in different directions.
Advantageously, the end points of the normal vectors having an identical length and being centered at an initial point are projected in thr
Feiereisen Henry M.
Picard Leo
Rodriguez Paul
Siemens Aktiengesellschaft
LandOfFree
Method for rendering, evaluating and optimizing a surface... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for rendering, evaluating and optimizing a surface..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for rendering, evaluating and optimizing a surface... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3279907