doi:10.1016/j.measurement.2006.05.019
Copyright © 2006 Elsevier Ltd All rights reserved.
Geometric error filtering performance of a cylindrical capacitive sensor (CCS) with circumferential gap
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Hyeong-Joon Ahna,
,
, Jong-Min Parkb and Dong-Chul Hanc
aSoongsil University, 511, Sangdo-dong, Dongjak-gu, Seoul 156-743, Republic of Korea
bKorea Air Force Academy, Ssangsu-ri, Namil-myun, Chungwon-Kun, Chongbuk 363-849, Republic of Korea
cSeoul National University, San56-1, Shinlim-dong, Gwanak-gu, Seoul 151-744, Republic of Korea
Received 23 October 2005;
revised 7 April 2006;
accepted 25 May 2006.
Available online 10 June 2006.
Abstract
In measuring the radial error motion, geometric errors of the target rotor can be filtered effectively using a cylindrical capacitive sensor (CCS). Since the CCS consists of several sensor electrodes, there are circumferential gaps between sensor electrodes for electric insulation. However, CCSs, so far, have been analyzed ignoring the circumferential gaps. In this paper, the electric potential distribution and capacitance in the circumferential gap are investigated through FE (finite element) analysis. Based on the FE analysis, extra capacitance in the circumferential gap is represented by the equivalent length extension of the sensor electrode so that we can investigate the circumferential gap effect analytically. The circumferential gaps are shown to affect significantly the geometric error filtering performance of the CCS. In addition, the robust analysis shows that the four-segment CCS is more robust to the uncertainties of the circumferential gap than the eight-segment CCS. Finally, using the beat phenomenon of a magnetic bearing rotor system, an odd harmonic geometric error is verified experimentally to be filtered effectively by selecting a particular CCS geometry considering circumferential gaps.
Keywords: Displacement measurement; Capacitive sensor and rotating machinery
Fig. 1. The CCS construction: (a) 3-D cross section; (b) construction details.
Fig. 2. Existing two CCS’s: (a) the four-segment CCS; (b) the eight-segment CCS.
Fig. 3. Simple model for guard gap corrections.
Fig. 4. FE model of CCS with the circumferential gap: (a) a CCS model for the FE analysis; (b) FE meshing results.
Fig. 5. Equipotential contour of the FE analysis results: (a) air and narrow gap; (b) epoxy and narrow gap; (c) epoxy and wide gap.
Fig. 6. Variations of the equivalent length extension due to various effects: (a) effect of the sensor thickness and the circumferential gap width; (b) effect of the air gap and the epoxy relative permittivity.
Fig. 7. Schematic diagram of the CCS with circumferential gaps.
Fig. 8. EAF variations according to circumferential gap widths.
Fig. 9. Analytical orbit measurement simulation of CCSs with narrow and wide circumferential gaps: (a) 1 mm circumferential gap; (b) 3.3 mm circumferential gap.
Fig. 10. Analytical orbit measurement simulation of the test rotor: (a) with the uncompensated CCS; (b) with the compensated CCS.
Fig. 11. Comparison of the expected harmonic error magnitudes.
Fig. 12. The experimental setup: (a) schematic diagram of the experimental setup; (b) the flexible rotor test rig; (c) the CCS housing.
Fig. 13. Fast beat phenomenon: (a) orbit plot; (b) time plot.
Fig. 14. The experimental results of orbit measurement of the test rotor: (a) with the uncompensated CCS; (b) with the compensated CCS.
Fig. 15. Fourier transformation of the experimental results: (a) large radius orbit in X direction; (b) large radius orbit in Y direction; (c) small radius orbit in X direction; (d) small radius orbit in Y direction.
Fig. A.1. Variables for the CCS mathematical model: (a) variables for a rotor and a sensor; (b) variables for a rotor position with roundness errors.
Fig. A.2. The EAF of the eight-segment CCS.
Fig. B.1. Test target rotor: (a) shape; (b) magnitudes of harmonic geometric errors.

Corresponding author.