Open Access
Add to BibSonomyAbstract
A stable dynamic detection scheme for magnetostrictive fiber-optic interferometric sensors is studied. The working principle is presented and the experimental verification is performed. The results show that the system employing the dynamic detection scheme has better stability and sensitivity in comparison with the system employing quadrature control technique.
©2006 Optical Society of America
1. Introduction
Magnetostrictive fiber-optic interferometric sensors have great advantages in detecting low-intensity magnetic field and significant progress has been made recently [1–5]. System stability is a key issue for practical applications [6–8]. So far, close-loop electronics is usually employed to keep the interferometer in quadrature state for the most sensitive detection. Such scheme utilizes adaptive phase compensation with a piezoelectric transducer (PZT) in the reference arm. The servo loop maintains the quadrature within milli-rad [6], which results in a minimum detectable phase shift of about micro-rad. However, quadrature condition may be significantly affected by environment parameters, such as temperature drift and mechanical and vocal vibration, which results in system instability, especially in weak signal detection.
This letter presents a stable dynamic detection scheme. Instead of close-loop phase locking, we dynamically detect the peak-to-peak value of the output voltage by applying a periodical triangular function on the piezoelectric transducer (PZT). This scheme can effectively eliminate the environmental disturbance. In the following, we analyze the principle and present experiment and explanations of the results. Finally, conclusion is given.
2. Principle
Michelson fiber-optic interferometer with two Faraday rotator mirrors can effectively overcome the polarization-induced fading [9, 10]. We deployed such interferometer in our experiment as shown in Fig.1. An optical circulator was used at the input of the interferometer so that the two orthogonal output signals from the coupler were differentially detected.
Fig. 1. Diagram of experimental setup. S: light source (laser), IS: isolator, C: circulator, FC: fiber coupler, RM: reference arm, SM: sensing arm, FRM: Faraday rotator mirror, P: photodiode, ω: function generator, DC: DC bias, SOL: solenoid.
The output signal detected by the photodiode can be expressed by [11]
where, R is the response of the photodetector with units of amperes per watt. I0 is the average optical power of the two arms of the interferometer. V is the visibility. ϕe represents the phase bias introduced by the PZT. It also includes the low frequency drift caused by the environmental effect such as temperature and vibration. Δϕ represents the phase shift induced by the low-intensity magnetic field to be measured, Hd, and the dither field, hcosωt, that enhances the system sensitivity by working at the transducer’s mechanical resonant frequency[11].
It is known that the detected signal depends strongly on the phase bias. In the conventional detection scheme, the quadrature state is achieved by feedback electronics [7, 8]. The servo loop needs a small error signal to maintain quadrature, which limits the system sensitivity for ultra-small signal detection. And the servo loop is also easy to be unlocked by large interference such as mechanical vibration. In order to overcome the shortage, we implemented a dynamical detection scheme. Instead of the close loop feedback, the signal applied to the PZT is a periodical triangular function. The amplitude of the periodical function should be large enough in order to produce a phase shift larger than π, so that the system experiences at least one quadrature condition. Meanwhile, the period of the function should also be properly chosen.
The differential output voltage can be expressed as [11]:
where G is the system gain. D is the phase-to-voltage conversion factor of PZT. n is the effective refraction index of the fiber core and L is the fiber length attached to the magnetostrictive materials. λ is the optical wavelength. C is the magnetostrictive parameter of materials. ξ is the elasto-optic correction factor[11]. Vtri(t) is the triangular function:
with A representing the amplitude and T the period of the triangular function.
The highest sensitivity, i.e, quadrature condition, is achieved when the PZT phase shift is equal to (2m+1)π/2 (m=0,1,2,…,). Figure.2 shows the output signal after the band pass filter (BPF) centered at ω. The low-intensity magnetic field can be detected by demodulating the peak-to-peak value, Vp-p, of the output voltage as [11,12],
It can be seen that Vp-p is proportional to the low-intensity field to be detected. In such dynamic detection scheme, any environmental phase drift will only cause the wave shifted in the X-axis and have no effect on the amplitude. Therefore environmental interference can be effectively eliminated and the system stability and sensitivity can be enhanced.
3. Experiment results
A cylindrical transducer based on Michelson interferometer was implemented by winding six layers of magnetostrictive material (a kind of Metglas ribbon) onto a bakelite framework and then coupling a sensing arm of 15 m single-mode fiber onto the ribbon with epoxy. The reference fiber was wound on the PZT and was put into the punch of the bakelite framework to make the transducer more compact. The difference of the fiber length of the two arms was controlled within 0.2 mm with the help of a precision reflectometer (HP8504). The experimental setup is shown in Fig.1. We compared the long-term stability between the quadrature control technique and the dynamic detection scheme. Figure.3 shows the results. It can be seen that unlocking takes place frequently in the system using the quadrature control technique due to saturation of the feedback integrator under relatively larger fluctuation. The system should be reset in order to return to normal state. Such problem entirely does not occur in our scheme.
As mentioned above, the amplitude of the triangular function should be large enough. Figure.4 shows the output Vp-p as a function of the amplitude of triangular function at T=1.4 ms. It can be seen that when the amplitude is larger than 19V, Vp-p maintains a constant value of about 1.0V. It indicates that the phase shift induced by the triangular function is more than π and therefore the interferometer always has a quadrature condition. For small amplitude, the quadrature condition can’t be assured and Vp-p varies randomly due to the environmental fluctuation.
Fig. 4. The output VP-P as a function of the amplitude of the triangular function with the period of 1.4ms.
The dependence of Vp-p on the period, T, of the triangular function was also investigated under the amplitude of 21.15V, and the result is shown in Fig.5. There is an optimal range for the period. T should be long enough to contain as much dither field periods as possible for reliable detection, i.e. 2π/T≪ω (the dither frequency in our experiment is about 19.6 kHz with amplitude h=2 µT). On the other hand, too long period of the triangular function may cause the output distortion due to the low-frequency noise.
Fig. 5. The output VP-P as a function of the period of the triangular function with the amplitude of 21.15V.
The low-intensity magnetic field responses employing the two schemes are shown in Fig.6. The triangular function is with amplitude of 20V and period of 1.4ms. It is shown that both scheme has linear response over large dynamic range. The sensitivity of the our scheme is about 8 percent higher than that of the quadrature control technique.
Though we used the fiber-optic Michelson magnetostrictive transducer as an example to show the feasibility of dynamic detection, the scheme is applicable to other kinds of fiber-optic interferometric transducers, including the one employing Mach-Zehnder interferometer. As is known, polarization-induced fading is a problem difficult to be overcome in a Mach-Zehnder interferometer.
4. Conclusion
In conclusion, we studied a stable dynamic detection scheme. Analysis is given and experimental verification is performed using a polarization-insensitive fiber-optic Michelson magnetostrictive transducer. The results show that, by appropriate setting of the amplitude and period of the periodical triangular function, the system employing the dynamic detection scheme has better stability and sensitivity over the one employing quadrature control technique. Moreover, the detection electronics is much simple and software detection can be implemented to make the system more compact and reliable.
Acknowledgments
This work is supported by the Foundation of Aeronautic Science, China (ID 05I57005) and NSFC (ID: 60377013, 90204006, 60507013).
References and links
1. Culshaw, “Optical Fiber Sensor Technologies: Opportunities and-Perhaps-Pitfalls,” IEEE J. Lightwave Technol. 22, 39–50 (2004). [CrossRef]
2. A. D. Kerseye and A. Dandridge, “An Overview of the Navy Fiber Optic Sensor Program,” in Proceedings of IEEE Conference on Lasers and Electro-Optics Society Annual Meeting(LEOS’90), 180–185 (1990).
3. F. Bucholtz, C. A. Villarruel, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, A. R. Davis, S. S. Patrick, K. P. Koo, K. G. Wathen, A. Dandridge, G. Wang, T. Lund, and H. Valo, “Fiber-Optic Magnetometer system for under sea applications,” Electron. Lett. 29, 1032–1033 (1993). [CrossRef]
4. A. D. Kersey and A. Dandridge, “Application of Fiber-Optic Sensors,” IEEE Trans. Compon. Hybrids Manuf. Technol. 13, 137–143 (1990). [CrossRef]
5. A Yariv and H.V. Winsor “Proposal for detection of magnetic fields through magnetostrictive perturbation of optical fubers” Opt. Lett. 5, 87–89 (1980) [CrossRef] [PubMed]
6. F. Bucholtz, C. A. Villarruel, A. R. Davis, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, S. S. Patrick, K. P. Koo, Gunnar Wang, Hilde ValØ, Terje Lund, A. G. Andersen, Robin Gjessing, E. J. Eidem, and Tor Knudsen, “Multichannel Fiber-Optic Magnetometer System for Undersea Measurements,” IEEE J. Lightwave Technol. 13, 1385–1394 (1995). [CrossRef]
7. A. D. Kersey, D. A. Jackson, and Michael Corke, “Single-Mode Fiber-Optic Magnetometer with DC Bias Field Stabization,” IEEE J. Lightwave Technol. 3, 836–840 (1985). [CrossRef]
8. D. M. Dagenais, F. Bucholtz, K. P. Koo, and A. Dandridge, “Detection of Low-Frequency Magnetic Signals in a Magnetostrictive Fiber-Optic Sensor with Suppressed Residual Signal,” IEEE J. Lightwave Technol. 7, 881–887 (1989). [CrossRef]
9. A. D. Kersey, M. J. Marrone, and M. A. Davis, “Polarization-insensitive Fiber Optic Michelson Interferometer,” Electron. Lett. 27, 518–520 (1991). [CrossRef]
10. M. J. Marrone, A. D. Kersey, C. A. Villarruel, C. K. Kirkendall, and A. Dandridge, “Elimination of Coherent Rayleigh Backscatter Induced Noise in Fiber Michelson Interferometers,” Electron. Lett. 28, 1803–1804 (1992). [CrossRef]
11. C. K. Kirkendall and A. Dandridge, “Overview of high performance fiber-optic sensing”. J. Phys. D: Appl. Phys. 37, 197–216 (2004). [CrossRef]
12. D. Y. Kim, H. J. Kong, and B. Y. Kim, “Fiber-Optic DC Magnetic Field Sensor with Balanced Detection Technique,” IEEE Photonics Technol. Lett. 4, 945–948 (1992). [CrossRef]
