Precise gouging-free tool orientations for 5-axis CNC machining☆
Introduction
Free-form NURBS surfaces are the de facto industry standard representation for 3D modeling and have been widely used in a variety of applications. Consequently, CNC machining of NURBS surface has been studied extensively in the past few decades. 5-axis machining, which provides two additional degrees of freedom, compared to 3-axis machining, attracted much attention due to the flexibility of 5-axis tool path planning, in contrast with the difficulties that 5-axis tool path generation entails.
There are two major technical challenges in 5-axis machining. One is how to optimally orient the tool so that it can approximate the local target surface properly. The other challenge is how to avoid collisions and gouging into the target surface, , and other check surfaces, or other surfaces of the object. Collision and gouging1 avoidance in 5-axis machining is more challenging compared to other applications, because the tool tip is typically in a tangential contact with at the milling contact point, denoted .
Consider 5-axis machining planning using a flat end cylindrical tool, , having a bottom circle, , as a cutting edge.2 For each contact point , two orientation degrees of freedom must be determined for , while maintaining tangential contact between and at . One natural approach to determine the orientation degrees of freedom uses a 2nd order approximation, to examine the osculating circle of a planar section of using the plane through . From Meusnier’s theorem [1], and using the two orientation degrees of freedom, one can attempt to match the curvature of and this planar section of at , which is also known as curvature matched machining [2], [3], [4], [5], [6], [7]. Curvature matched machining provides a 2nd order approximation and is a simple way to define the tool orientation, and it has been widely adapted in tool path computation. However and as a side effect of curvature matched machining, is likely to locally gouge into , near , because the planar section curve of typically has an increasing or decreasing curvature at , while presents a constant curvature (see Fig. 1(a)).
Hyper-osculating circles (HOCs) [8], [9] alleviate this difficulty and vastly reduce the possibility of gouging into . Based on a 3rd order differential analysis, HOCs share the same position, tangent direction, curvature, and curvature derivative (which is zero for a circle), with ’s planar section at . In other words, an HOC is an osculating circle that is located at a curvature extreme point of the planar section of and thus it resolves the local potential gouging of into (see Figs. 1(b) and 5). However, this 3rd order approximation also has a limitation. Since the HOCs should satisfy both curvature and curvature derivative constraints, typically only a limited number of candidate configurations exist. As a consequence, a hyper-osculating configuration is not always feasible, and even less so when various constraints in practical situations are imposed, e.g., angular orientation limits on the CNC machine, global collisions, etc.
To help overcome this limitation of the HOCs, we also consider cases of two-tangential contacts between and (see Fig. 1(c)). Henceforth and unless otherwise stated, ‘contact’ will denote a tangential contact. The key idea of our approach is based on the fact that there always exist two-contact configurations in the local neighborhood of hyper-osculating configurations.3 From this observation, we can augment the HOCs and switch to nearby two-contact configurations whenever necessary, while providing good approximation quality in terms of the curvature difference between the bottom circle and the surface section at and being gouging-free.
The rest of this paper is organized as follows. In Section 2, we briefly review previous related work. Section 3 introduces HOCs and establishes the algebraic conditions for two-contact configurations between and . Then, we propose a global optimization algorithm for the tool path that maximizes the approximation quality of the HOCs and the two-contact configurations. Experimental results are reported in Section 4, and the paper is concluded in Section 5.
Section snippets
Related work
Related work on 5-axis milling mostly appeared in the CAD and mechanical engineering literature (see reference works [10], [11]). There are numerous papers contributing geometric methods to freeform surface CNC machining, including multi-axis (4- and 5-axis) machining. While this work focuses on contact analysis at the tip of the tool, in 5-axis machining, side or flank milling is one method that attracted much attention. See a recent survey in [12] on flank milling.
Numerous efforts were made
The algorithm
Consider a continuous target surface and let and be the partial derivatives and the normal field of . Now consider one tool path in , , of a flat end tool with a bottom circular edge , and let be the current contact point. Further let, be the radius of , the center point of and the plane containing . Finally, let be the intersection curve between plane and ; see Fig. 3.
Recall that
Experimental result
In this section, we present results of using the tool contact planning algorithms on several free-form surfaces. Fig. 10 illustrates the limitations when only HOCs are considered. In order to fairly evaluate the potential of HOCs, we computed optimized contact paths with a level set approach (see figure caption). Even then, the coverage of the surface by HOC strips is too low.
In the coming examples, the optimization algorithm computes 200 samples along the tool path and 200 samples along ,
Conclusion and future work
The proposed optimization algorithm naturally combines HOCs and two-contact configurations by treating the hyper-osculating case as a special type of the two-contact configuration, and by simply minimizing the curvature difference between and . The tool path orientations generated by the global optimization procedure tend to maintain the hyper-osculating configurations when it is accessible, and smoothly switch to nearby two-contact configurations if there is no hyper-osculating position or
Acknowledgments
This work was supported in part by the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme FP7/2007–2013/ under REA grant agreement PIAP-GA-2011-286426, and was supported in part by the Israel Science Foundation (grant No. 278/13).
References (33)
- et al.
Locally optimal cutting positions for five-axis sculptured surface machining
Comput Aided Des
(2003) - et al.
Curvature catering—a new approach in manufacture of sculptured surfaces (part 1. Theorem)
J Mater Process Technol
(1993) - et al.
Curvature catering—a new approach in manufacture of sculptured surfaces (part 2. Methodology)
J Mater Process Technol
(1993) - et al.
5-axis flank milling: a state-of-the-art review
Comput Aided Des
(2013) - et al.
Cutter-location data optimization in 5-axis surface machining
Comput Aided Des
(1993) - et al.
New numerical algorithms to optimize cutting operations of a five-axis milling machine
Appl Numer Math
(2004) - et al.
Optimizing tool orientations for 5-axis machining by configuration-space search method
Comput Aided Des
(2003) - et al.
Intersection approach to multi-point machining of sculptured surfaces
Comput Aided Geom Design
(1998) - et al.
Multi-point tool positioning strategy for 5-axis mashining of sculptured surfaces
Comput Aided Geom Design
(2000) - et al.
Comparison between multi-point and other 5-axis tool positioning strategies
Int J Mach Tools Manuf
(2000)
Correlation between machining direction, cutter geometry and step-over distance in 3-axis milling: application to milling by zones
Comput Aided Des
Tool path generation for freeform surface models
Comput Aided Des
Space-filling curves in tool-path applications
Comput Aided Des
A classified bibliography of literature on NC tool path generatio
Comput Aided Des
Algorithms for collision detection and avoidance for five-axis NC machining: a state of the art review
Comput Aided Des
Differntial geometry of curves and surfaces
Cited by (0)
- ☆
This paper has been recommended for acceptance by Dr. Vadim Shapiro.