The surface roughness of six vanes were measured on both concave and convex sides, and tabulated in Table 2. The average roughness ranges from
1.022 to 1.4 microns Ra, better than the required 1.6 microns Ra.
Vanes before and after blending are shown in Figure 20. A very smooth airfoil profile was achieved by the SMART system. Further visual
inspection shows no visible transition lines from the non-brazed area to the brazed one, no visible blending marks in the cutting path overlap areas, and no burning marks. The curvature transition from the concave to convex airfoil is very smooth and more consistent than the one generated by manual blending.
Table 2 Roughness measurement results.
Vane no.
1 2 3 4 5 6
i?a(um) Convex 1.378 1.059 1.228 1.003 1.197 1.196
Concave 1.417 1.090 1.086 1.041 0.936 0.980
Average Ra
(urn) 1.400 1.075 1.157 1.022 1.067 1.088
Figure 20 Vanes before (left) and after (right) robotic grinding and polishing.
It is worth noting that the consistency of finish profiles also benefits the downstream laser drilling operation. If the finish profile is not consistent in terms of wall thickness, airfoil shape and leading edge position, the cooling holes generated by the laser machine may converge, and affect the aerodynamic performance of the engine, and in many cases be rejected by Quality Assurance. In the automated system, such failure is minimised, if not eliminated.
7.3 Wall Thickness
Following the existing practice in the aerospace overhaul industry, the wall thickness needs to be quantified. Minimum wall thickness must be maintained in order not to undermine the airfoil strength. In the selected case, seven points are measured against the minimum wall thickness, points
#1 to #7 in Figure 21.
Figure 21 Positions of checking points.
Figure 22 plots the minimum wall thickness, and the measurement results for three vanes. It shows that all measurements at the designated seven points are above the minimum wall thickness. Figure 23 shows the samples of sectioned airfoils. Trailing edge thickness is maintained, and the smooth curvature of leading edge is confirmed.
0.12 T 1
0.02
0 J , , , , , 1
1 2 3 4 5 6 7
Checking Point
Figure 22 Results of wall thickness measurement.
Figure 23 Cross sections of polished vanes: leading edge (left) and trailing edge (right).
8. Concluding Remarks
In a concerted effort, we have successfully developed a Knowledge-Based Adaptive Robotic System for 3D Profile Grinding and Polishing. Around the finishing robot, Self-Aligned End-effector (SAE), Passive Compliant Tools (PCT) and In-Situ Profile Measurement (IPM) system have been developed. Template-based Optimal Profile Fitting (OPF) algorithm, and Adaptive Robot Path Planner (ARP) have been developed to overcome the part distortions inherent in component overhaul and to satisfy force and compliance control required for robotic blending. The technological modules have been built into the first working prototype SMART 3D Grinding/Polishing System.
The synergistic combination of hardware and software solutions enables the SMART system to meet stringent quality requirements and design criteria, such as profile smoothness, surface roughness, leading edge transition and height, minimum wall thickness, removal of transition lines between brazed and non-brazed areas, etc. The SMART system, the first of its kind for blending distorted airfoils, has been benchmarked against critical quality measures, and satisfies all product and process requirements.
It shortens the cycle time from 10 minutes (manual blending) to an average of 5.75 minutes, resulting in an improvement of 42.5%. It has since been installed for production use.
The technological breakthrough lays a strong foundation for future explorations of robotic machining for other advanced applications, such as overhauling fan blades (currently done by operator-assisted CNC), and manufacturing new aircraft components. Another challenge is to automate the finishing process of precision mechanical components such as 3D
moulds, which is still accomplished manually despite earlier attempts by various schools worldwide.
References
1. Engel, T. and Tomastik, R. "Description of the chamfering and deburring end- of-arm tool (CADET)," The Fourth International Conference on Control, Automation, Robotics and Vision, Singapore, 2-6 December 1996, pp. 629 - 633
2. Berger, U., Janssen, R. and Brinksmeier, E. "Advanced mechatronic system for turbine blade manufacturing and repair", International Conference on Computer Integrated Manufacturing, Singapore, 21-24 October 1997, pp. 1395-1404.
3. Chen Y.H., and Hu, Y.N. "Implementation of a robot system for sculptured surface cutting. Part 1: rough machining", International Journal of Advanced Manufacturing Technology (1999) 15, pp. 624-629.
4. Hu, Y.N., and Chen, Y.H. "Implementation of a robot system for sculptured surface cutting. Part 2: finish machining", International Journal of Advanced Manufacturing Technology (1999) 15, pp. 630-639.
5. Ozaki, F., Jinno, M., Yoshimi, T., Tatsuno, K., Takahashi, M., Kanda, M., Tamada, Y., and Nagataki, S. "A force controlled finishing robot system with a task-directed robot language," Journal of Robotics and Mechatronics, 1995, Vol. 7, No. 5, 1995, pp. 383-388.
6. Kunieda, M. and Nakagawa, T. "Robot-polishing of curved surface with magneto-pressed tool and magnetic force sensor", Proceedings of 25th
International MTDR Conference, April 1985, pp. 193-200.
7. Woon, L.C., Ge, S.S., Chen, X.Q., and Zhang, C. "Adaptive neural network control of coordinated manipulators", Journal of Robot Systems, April 1999, Vol. 16, No. 4, pp. 195-211.
8. Lancaster, P., and Salkauskas, K. "Curve and Surface Fitting - An Introduction," Academic Press Inc. (London) Ltd., London, 1986.
9. Chapra, S.C., and Canale, R.P. "Numerical Methods for Engineers - With Programming and Software Applications," Third Edition, WCB/McGraw-Hill, Boston, 1998.
10. Press, W.H., Flannery, B.P., Teukolsky, S.A., and Vetterling, W.T.
"Numerical Recipes in C," Cambridge University Press, New York, 1988.
11. Reklaitis, G.V., Ravindran, A., and Ragsdell, K.M. "Engineering Optimization: Method and Applications," John Wiley & Sons, Inc., New York, 1983.
\2. Hooke, R. and Jeeves, T.A. "Direct search of numerical and statistic problems," J. ACM, 1996, Vol. 8, pp. 212-229.
13. Gong, Z.M., Chen, X.Q. and Huang, H. "Optimal profile generation in distorted surface finishing", IEEE International Conference on Robotics and Automation, San Francisco, USA, 22-24 April 2000, pp. 1557-1562.
14. Burden, R.L. and Faires, J.D. "Numerical Analysis," Fifth edition, PWS Publishing Company, Boston, 1993.
15. Pizer, S.M. "Numerical Computing and Mathematical Analysis," Science Research Associates, Chicago, 1975.
16. Gellert, W. and Kustner, H. "The UNR Concise Encyclopedia of Mathematics," Second edition, Van Nostrand Reinhold Company, New York, 1989.
17. Fu, K.S., Gonzalez, R.C. and Lee, C.S.G. "Robotics: Control, Sensing, Vision, and Intelligence," McGraw-Hill Book Company, New York, 1987.
18. Craig, J.J. "Introduction to Robotics: Mechanics and Control," Second edition, Addison-Wesley Publisher Company, Mass, 1989.
19. Chen, X.Q., Gong, Z.M., Huang, H., Ge, S.S. and Zhu, Q. "Development of SMART 3D polishing system" Industrial Automation Journal, A Singapore industrial automation association publication, April-June 1999, pp. 6-11.
20. Yamaha Finishing Robot - User's Manual Version 1.4.