- Email: [email protected]
Alternative interrogation method for a dual laser sensor based on fiber Bragg gratings to measure temperature using the fundamental beating frequency intensity
Optics & Laser Technology 67 (2015) 159–163
Contents lists available at ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
Alternative interrogation method for a dual laser sensor based on fiber Bragg gratings to measure temperature using the fundamental beating frequency intensity O. Méndez-Zepeda, S. Muñoz-Aguirre n, G. Beltrán-Pérez, J. Castillo-Mixcóatl Facultad de Ciencias Físico-Matemáticas, Benemérita Universidad Autónoma de Puebla, 18 sur y Av. San Claudio, Col. San Manuel, CU, C.P. 72570 Puebla, México
art ic l e i nf o
a b s t r a c t
Article history: Received 20 August 2014 Received in revised form 11 October 2014 Accepted 16 October 2014 Available online 7 November 2014
In this work we present an alternative interrogation method for a dual sensor based on a fiber laser used to measure temperature in two remote locations simultaneously. The dual laser consisted of two Fabry– Perot cavities each conformed by two fiber Bragg gratings (FBG). For each cavity, one FBG was used as the reference and the other one as a sensing element. The sensing element interrogation was performed by the quantification of the fundamental beating frequency (FBF) intensity, which was calculated using the fast Fourier transform algorithm. The laser emissions were centered at 1549 and 1556 nm, while the lengths of cavities were of 300 and 400 m, which corresponds to FBFs of 334 and 258 kHz, respectively. The quantification of the temperature was calculated from the difference between the FBF values of both cavities. Such difference describes a geometrical plane in function of the two sensing FBGs temperatures. Consequently, it was possible to achieve temperature measurements in a range of 25–28 1C for the two sensors. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Fiber Bragg grating Fundamental beating frequency Optical fiber laser sensor
1. Introduction In general, the sensors based on fiber Bragg gratings (FBG) require methods or techniques for measuring the spectral shifts that the Bragg wavelength (λB) presents when the FBG is in contact with some physical variable such as strain or temperature variations. Then, in order to detect and measure the changes of such variables through spectral shifts it is required expensive equipment such as the optical spectrum analyzer (OSA) [1], electrical spectrum analyzer (ESA) [2], tunable lasers [3], and different systems such as Sagnac filters [4,5], tunable Fabry–Perot filters [6], Mach–Zehnder interferometers [7,8], among others. Therefore, it is required to develop detection and measurement techniques that are economical and easy to implement. On the other hand, the measurement of two or more variables with the same setup using FBGs becomes difficult and expensive, although the development of some sensors that perform physical variables measurements through the quantification of FBGs reflection or transmission spectra shifts have been reported. For instance, some works use the time division multiplexing (TDM) [9], arrayed waveguide gratings (AWGs) demultiplexer [10], while
n
Corresponding author. Tel.: þ 52 222 2295500x2176; fax: þ52 222 2295636. E-mail addresses: [email protected], [email protected] (S. Muñoz-Aguirre). http://dx.doi.org/10.1016/j.optlastec.2014.10.014 0030-3992/& 2014 Elsevier Ltd. All rights reserved.
another employs the wavelength-division multiplexing (WDM) technique [11] or in the case of Lu et al. that report a simultaneous discrimination of axial strain and temperature FBGs through the wavelength shift [12]. In general, all the works above mentioned use complicated experimental setups. Other laser sensors based on FBGs cavities involve the use of the determination of the fundamental beating frequency (FBF) shifts to quantify the physical variables. For instance, a sensor principle based on the measurement of the beating frequency shifts, which are provoked by birefringence changes, has been reported. In this category we can find the polarimetric fiber laser sensors, which require two orthogonal polarization modes to generate polarization mode beating as the sensing signal [13–16]. They focus mainly in the determination of the FBF shifts. In a previous work we have reported that there exists a relationship between the output laser intensity and the overlapping region of two FBGs reflection spectra caused by one FBG stretching as result of a physical variable [17]. It was also found that the output optical power variations were linear in relation to the sensing FBG stretching. However, if this setup is used to evaluate more than one variable simultaneously, it would not be possible to discriminate the variables through the optical power intensity, since the total response is the combination of both intensities; if there is an intensity variation it would not be possible to know which cavity is the one that provokes such variation. The cavity discrimination would be possible if the output optical power of the dual laser were analyzed in the frequency domain, in such case,
160
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
the response of each sensor would be identified from the FBF corresponding to each cavity. The last affirmation is based on the fact that the length of a laser cavity determines the beating frequencies between any pair of resonance modes inside the cavity that can be calculated from Equation (1).
νq ¼ q
c 2d
q ¼ 1; 2; 3; …
ð1Þ
where c is the light speed inside the cavity and d is the cavity length. For the particular case when q¼ 1 (beating between two adjacent modes), the FBF is given by Equation (2). FBF ¼ Δν ¼
c 2d
ð2Þ
From here, if two cavities with different length are used, it can be expected that each laser emission can be identified by their FBF value according to Equation (2). On the other hand, if the dual laser output is analyzed in the frequency domain, each FBF intensity change as a function of the laser power intensity which depends on the FBGs spectra overlapping. It is important to mention that the quantification and identification using the FBF intensity has not been used so far and that its implementation would decrease cost and make the interrogation setup simpler, especially when the system is used to measure two variables simultaneously. In this work we propose an alternative simple method to interrogate a dual temperature sensor, where each sensor was set in a different location. The interrogation was performed by the determination of the FBF intensities of a multilongitudinal dual laser, which has a few advantages for sensing applications over the above mentioned schemes. First, the laser cavities were constructed with commercial elements, avoiding the use of special FBG written in the erbium doped fiber (EDF) and the polarization states control that are quite sensitive to fiber bends and twists, which must be avoided especially when the physical variable produces small changes in the birefringence. Secondly, a large cavity provides much higher laser power than single-mode lasers, and hence a higher SNR can be obtained. Furthermore, our scheme can be used as a dual and remote sensor since the standard fiber that conform part of the cavity can be installed far away from the system. On the other hand, the FBF can be measured with a conventional data acquisition (DAQ) card, without using expensive equipment such as the optical or electrical spectrum analyzers (OSA and ESA, respectively).
2. Experimental The experimental setup is shown in Fig. 1. The dual fiber laser consisted of two Fabry–Perot cavities with lengths of 300 and 400 m, respectively, constructed with FBGs and standard optical fiber SMF28 (Corning). Both cavities share 7 m length of EDF as the gain medium.
These lengths are important since the fiber can be installed far away from the interrogation system and it would allow using the sensors in two remote locations. Furthermore, the FBF values of long cavities are in the order of hundreds of kHz, which is not so high frequency and it can be measured using conventional electronic circuits. The gain medium was pumped by a laser diode (26-8052-100, JDSU) with a wavelength of 980 nm, which was adjusted to an output optical power of 35.5 mW and was coupled to the EDF through a wavelength division multiplexer (WDM). Each cavity was constructed with two FBGs (Bragg Photonics Inc.), one of them was the reference element (FBG-R) and the other one was the sensing element (FBG-S). The two reference elements were FBG-R1 (λB ¼1549 nm, reflectivity of 58%) and FBG-R2 (λB ¼1556 nm, reflectivity of 58%), while the sensing elements were FBG-S1 (λB ¼ 1546 nm, reflectivity of 58%) and FBG-S2 (λB ¼1553 nm, reflectivity of 59%). Without any excitation all the FBGs λB are different and there is not laser emission since there is not conformed any cavity. However, when the spectrum of the FBG-S starts overlapping over that of FBG-R, provoked by a physical variable, the cavity is set and the laser intensity starts and it changes as a function of the spectral overlapping of the FBGs in each cavity. Moreover, if we try to measure temperature with this FBGs, we need a very high temperature variations, since the spectral shifts of the FBGs are quite slight (approximately 10 pm/1C). For this reason, the system needs an initial condition of spectra overlapping which results in an initial dual emission that is reached when we apply a stretching on the sensing FBG through a micrometer screw, which is not shown in the scheme of Fig. 1 since it was used only to set such initial condition. It is important to mention that this initial stretching does not affect the FBF, since it is a stretching of a few tens of micrometers in hundreds of meters of the total cavity length. After reaching this initial condition, we can start to measure temperature variations. The temperature of each FBG-S was controlled by a Peltier cell and only a portion of 4 cm of the fiber that contains the FBG was exposed to temperature changes. This was performed for each cavity. The signal conditioning electronic circuit was composed of a conventional photodiode (FGA10, detection range from 800 to 1800 nm) and a transimpedance amplifier, which provides an output voltage in the range from 1.2 to 4.8 V dc and a bandwidth of 500 kHz. These values were adequate for the frequency range that we were measuring in our system, whose maximum frequency was around 334 kHz. Finally, the obtained signal was digitalized using a DAQ card (PCI-1712, ADVANTECH) with a sampling rate of 1 MS/s and a 12-bit resolution. This signal was analyzed using the FFT algorithm to determine its spectrum in the frequency domain and then to determine the frequency value and amplitude of the FBF. The data acquisition procedure was as follows: a measurement of 222 samples (during approximately 5 s) was performed and the obtained data vector was segmented in sub-vectors of 28, since the FFT algorithm works better with data sets of 2n sizes. Then, the FFT was calculated in order to determine the FBF intensity for each segment. Finally, the behavior of the FBF intensity in function of the FBG-S temperature was studied.
Fig. 1. Experimental setup.
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
FBF intensity x 10-3 (a.u.)
10 FBF = 328 kHz 8 6 4 2 0 0
100
200 300 Frequency (kHz)
400
500
Fig. 2. FFT of the acquired signal for a single cavity.
1.0
FBF intensity x 10-3 (a.u.)
R2 = 0.9791 0.8 0.6 0.4 0.2 25
26 27 FBG-S temperature (°C)
28
Fig. 3. FBF intensity as a function of the FBG-S temperature for a single cavity.
3. Results and discussion First of all, the study of only one cavity was performed to measure and determine the FBF. Therefore, it was carried out a measurement of the sensor output signal during 256 μs (28 data) at a rate of 1 MS/s and the FFT algorithm was applied to these data. In Fig. 2, the spectrum in the frequency domain of the single sensor output signal is shown. A peak located in 328 kHz, which corresponds to the FBF for a 314 m cavity length is clearly observed. Such FBF value is according with the one calculated from Equation (2), which was 324 kHz. The FBF measurement resolution was of 3.9 kHz. As it was already mentioned above, it can be expected that the intensity of the FBF peak changes as a function of the spectra overlapping of the FBGs, which is originated by the temperature variations detected by the FBG-S. In order to corroborate this statement, the FBF intensity changes in relation to the temperature in a range of 25–28 1C in steps of 1 1C were performed for one cavity configuration. The result is shown in Fig. 3. The FBF intensity has a slight offset; since an initial emission was set in a suitable point where both FBG spectra were slightly overlapped. The temperature measurement was performed from this point. The FBF intensity has a linear tendency in function of temperature with a correlation factor of R2 ¼ 0.9791, which indicates a quite good correlation when a linear fitting was performed. This behavior is similar to the case when the relationship between optical power and the FBG spectra overlapping was studied (inset in Fig. 3) and it is in agreement with results obtained in a previous work [17]. This fact indicates that the FBF intensity has a direct relationship with the spectra overlapping area of both FBG that conform the cavity. For this reason, we can say that the alternative
161
method proposed in this work can be used as a measurement method of a physical variable that change the spectra overlapping of the FBG, which can be stretching or temperature changes. In this particular case, it was possible to measure temperature variations with sensitivity of 1.68 10 4 1C 1, and a maximum error of 6%, when various measurements were performed, as is shown by the error bars in Fig. 3. Such sensitivity depends on the spectra width and profile of the FBGs that compose the cavity. It is important to mention that the cavities are conformed by FBGs that has a spectral width of 1.2 nm, and due to the initial spectral overlapping we have a quite short range to measure temperature. However we are interested in showing an alternative interrogation method to measure two variables based on the quantification of the FBF intensity instead of FBF shifts, which requires expensive measurement equipment. The measurement range could be increased by changing the FBGs with wider reflection spectra, although a further study would be necessary. For the purpose of measuring two variables simultaneously (temperature in two remote locations), it is important to obtain a dual emission, which can be achieved with two cavities in the same setup (Fig. 1). As it was already mentioned above, it is not possible to identify which cavity is emitting or its emission intensity when the measurement is performed using just one photodiode to detect optical power changes. With the method proposed in this work, each emission can be identified in the frequency domain employing the corresponding FBF value for each cavity. It is important to remark that it is quite difficult to obtain a stable and simultaneous dual emission [18,19]. In our case, we could achieve such simultaneous and stable dual emission, whose peaks can be controlled in an independent way. An example of the dual emission measured with an OSA is shown in Fig. 4a. The two peaks can be clearly observed, whose wavelengths are centered at the λB of the corresponding FBG-R; i.e. at 1549 and 1556 nm, respectively. On the other hand, the corresponding FBF value for each emission was determined and it was found that those values were in agreement with the cavities lengths. In Fig. 4b, the FBF spectrum of both cavities is shown, where two peaks can be clearly observed centered at 258 y 334 kHz. From now on we will call them FBF1 and FBF2, respectively. According to Eq. (2), the frequency peaks correspond to cavity lengths of 400 and 300 m, respectively. With this, we show that it is possible to identify, in an independent way, the emissions of the cavities and analyze the behavior of the FBF intensities when the two FBG-S's are under the influence of a physical variable. In this case, we can measure temperature changes by using just one conventional photodiode. Once the dual emission was achieved, the behavior of the FBF1 and FBF2 intensities in function of the temperature changes of both FBG-S (FBG-S1 and FBG-S2) was analyzed. The measurement of the FBF intensity changes of both cavities was performed under the following conditions: First, the FBG-S1 was kept at a constant temperature while that of the FBG-S2 was swept from 25 to 28 1C in steps of 1 1C. This process was repeated for FBG-S1 temperatures of 26, 27 and 28 1C, respectively. The results of the measurement of both FBFs intensities for these conditions are shown in Fig. 5. For each case, we can observe an increment of the FBF2 intensity as the FBG-S2 temperature was increased. This increment is in agreement with the results shown in Fig. 3. On the other hand, one would expect that the FBF1 intensity remain constant for each FBG-S2 temperature sweep, however a decrement was observed. Such decrement is more evident for the case when FBG-S1 was at 28 1C (Fig. 5d). This could be due to the modes competition inside the cavities since both of them share the same gain medium and probably the gain increment of the cavity 2 would affect the gain of the cavity 1 and vice versa, as can be observed in Fig. 5, a and d. The same behavior was observed in the opposite case, that is, when the FBG-S2 was under a constant temperature while the FBG-S1
162
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
0
Laser emission1
Laser emission 2
1549 nm
1556 nm
10 FBF1 258 kHz
(a.u.) -3
-20
FBF intensity x 10
Optical power (dBm)
-10
-30 -40 -50
8 6 FBF2 334 kHz
4 2 0
1545
1550
1555
0
1560
100
200
300
400
500
Frequency (kHz)
Wavelength (nm)
Fig. 4. (a) Emission of the two cavities, centered at 1549 and 1556 nm, (b) FBF intensity corresponding to each emission in the frequency domain.
4 2 0
8
FBF1 FBF2
6 4 2 0
10 8 6 4 2 0
26 25 26 26 26 27 26 28 Temperature (°C)
25 25 25 26 25 27 25 28 Temperature (°C)
FBF1 FBF2
10
FBF intensity x10-3 (a.u)
6
10
FBF intensity x10-3 (a.u)
8
FBF1 FBF2
FBF intensity x10 -3 (a.u)
FBF intensity x10-3 (a.u.)
10
FBF1 FBF2
8 6 4 2 0
27 25 27 26 27 27 27 28 Temperature (°C)
28 25 28 26 28 27 28 28 Temperature (°C)
Fig. 5. FBF intensity changes as a function of temperature for the two cavities.
4 2 0 25 25 26 25 27 25 28 25 Temperature (°C)
8
FBF1 FBF2
6 4 2 0 25 26 26 26 27 26 28 26 Temperature (°C)
10 8
FBF1 FBF2
6 4 2 0 25 27 26 27 27 27 28 27 Temperature (°C)
10
FBF intensity x10-3 (a.u.)
6
10
FBF intensity x10-3 (a.u.)
8
FBF1 FBF2
FBF intensity x10 -3 (a.u.)
FBF intensity x10-3 (a.u.)
10
FBF1 FBF2
8 6 4 2 0
25 28 26 28 27 28 28 28 Temperature (°C)
Fig. 6. FBF intensity changes as a function of temperature for the two cavities.
temperature was swept in the range from 25 to 28 1C, as it is shown in Fig. 6. Another important aspect is that if we focus only in one of the FBFs intensities, for instance the FBF2, we can observe that the slope decreases as the FBG-S1 temperature increases. Such effect can be used to have a sensor with variable sensitivity in FBG-S2, which is controlled by the temperature of the FBG-S1. According to the above results, since both FBF intensities changed at the same time, it would be possible to find a relationship of the difference between both intensities (DFBF ¼FBF2 FBF1). For this reason, it was decided to characterize such behavior by calculating the differences of the FBF intensities for each pair of temperature values (T1 and T2) of the FBG-S1 and FBG-S2, respectively. The obtained results are shown in Fig. 7, where it can be observed that the DFBF describes a plane, which has an inclination in the direction of both independent axis, that correspond to the temperature of both FBG-S's T1 and T2. Therefore, it is possible to determine a pair of
temperature values for each DFBF. The equation that describes this plane was calculated from a fitting of the experimental measurements (Eq. 3), which had a correlation coefficient of 0.9830. Such correlation value is adequate to determine T1 and T2. DFBF ¼ –1:44 10 3 T1 þ 2:03 10 3 T2–13:84 10 3
ð3Þ
From Equation (3) is possible to quantify the temperature of each FBG-S, simultaneously. The temperature evaluation is as follows: First, the DFBF is calculated from the measured values FBF1 and FBF2. Then, the plane equation is evaluated by sweeping with slight increments all the possible temperature values in the range of 25– 28 1C for both FBG-S´s. At this point, when the calculated FBF difference equals the measured value, we can say that the temperature of each FBG-S's has been determined. Finally, from Equation (3), it is important to mention that the sensitivity depends on the temperature of both FBG-S. Since there are two sensors, two
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
163
Acknowledgments The present work was partially supported by the projects CONACYT Jóvenes Investigadores 61126, VIEP-BUAP CAMJ-EXC14I, BEPG-EXC14-I, and MUAS-EXC14-G.
References
Fig. 7. DFBF of the measured data and fitting of the experimental measurements (shadowed plane).
sensitivities defined by the temperature of each FBG-S, were obtained. In this case, we obtained a sensitivity of 1.44 10 3 1C 1 for the sensor 1 and 2.03 10 3 1C 1 for the sensor 2. It is clear that this sensitivity is defined for the DFBF. The temperature range that can be measured depends on the reflection spectra width of the FBGs, in our particular case, before starting the temperature measurement, it was applied an initial strain to the FBG-S with the purpose of adjusting the laser intensity in a value where it were possible to achieve the dual emission and to determine the FBF of both emissions, for this reason the dynamic range was limited to the above mentioned values. Such dynamic range probably can be extended if FBGs with wider reflection spectra were used, although a further study would be necessary to determine if both FBGs that compose each cavity must be wide or only one of them. This study is out of the scope of the present paper, since we are only interested in showing the concept of the dual sensor with two cavities working simultaneously. 4. Conclusions In this work, we presented an alternative method to measure temperature in two different locations simultaneously using a dual laser. It consisted of two Fabry–Perot cavities composed by FBGs that share a single gain medium. The setup has the advantage that it is quite cheap and employs conventional elements. The FBF intensities of the dual laser were determined by a data processing performed with a commercial DAQ and a computer program that uses the FFT algorithm. The results showed that the FBF intensity was proportional to the spectra overlapping of the FBGs that compose each cavity. Moreover, when two cavities were used, it was possible to measure temperature changes in two remote locations in the range of 25–28 1C by evaluating the difference between both FBF intensities (DFBF). It was found that DFBF defines a geometrical plane in function of the temperature of both FBG-S's. Finally, although DFBF depends on both FBF signals, the two sensors can be operated in an independent way with their own sensitivity, which in this case, was constant for the temperature of each FBG-S.
[1] Yasukazu Sano, Toshihiko Yoshino. Fast optical wavelength interrogator employing arrayed waveguide grating for distributed fiber Brag grating sensors. J Lightwave Technol 2003;21(1):132–9. [2] Shengchun Liu, Zuowei Yin, Liang Zhang, Liang Gao, Chen Xiangfei, Cheng Jianchun. Multilongitudinal mode fiber laser for strain measurement. Opt Lett 2010;35(6):835–7. [3] Chan CC, Gong JM, Jin W, Demokan MS. Investigation of unwanted interferometric signals in a fiber Bragg grating sensor using a tunable laser and a first derivative interrogation technique. Opt Commun 2000;173:203–10. [4] Xiufeng Yang, Chun-Liu Zhao, Qizhen Peng, Zhou Xiaoqun, Chao Lu. FBG sensor interrogation with high temperature insensitivity by using a HiBi-PCF Sagnac loop filter. Opt Commun 2005;250:63–8. [5] Kyung-Rak Sohn. Fiber Bragg grating-tuned feedback laser flow sensor system. Sensor Actuat A 2012;179:1–4. [6] Kersey D, Berkoff TA, Morey WW. Multiplexed fiber Bragg grating strainsensor system with a fiber Fabry–Perot wavelength filter. Opt Lett 1993;18 (16):1370–2. [7] Ya-nan Zhang, Yong Zhao, Wang Qi. Improved design of slow light interferometer and its application in FBG displacement sensor. Sensor Actuat A 2014;214:168–74. [8] Qiqi Yao, Hongyun Meng, Wei Wang, Hongchao Xue, Xiong Rui, Huang Ben, et al. Simultaneous measurement of refractive index and temperature based on a core-offset Mach–Zehnder interferometer combined with a fiber Bragg grating. Sensor Actuat A 2014;209:73–7. [9] Sun Q, Liu D, Wang J, Liu H, Xia L, Shum P. Experimental demonstration of multi-point temperature warning sensor using a multi-channel matched fiber Bragg grating. IEEE Photonic Technol Lett 2008;20(11):933–5. [10] Hongqiang Lia, Yang Lia, Enbang Lib, Xiaye Donga, Yaoting Baia, Yu Liua, et al. Temperature-insensitive arrayed waveguide grating demodulation technique for fiber Bragg grating sensors. Opt Laser Technol 2013;51:77–81. [11] Chang-Seok Kim, Tae Ho Lee, Yun Sik Yu, Young-Geun Han, Sang Bae Lee, Myung Yung Jeong. Multi-point interrogation of FBG sensors using cascaded flexible wavelength-division Sagnac loop filters. Opt Express 2006;14 (19):8546–51. [12] Lu P, Men L, Chen Q. Resolving cross sensitivity of fiber Bragg grating with different polymeric coatings. Appl Phys Lett 2008;92:171112. [13] Kringlebotn JT, Loh WH, Laming RI. Polarimetric Er3 þ -doped fiber distributed-feedback laser sensor for differential pressure and force measurements. Opt Lett 1996;21(22):1869–71. [14] Bai-Ou Guan, Long Jin, Yang Zhang, Hwa-Yaw Tam. Polarimetric heterodyning fiber grating laser sensors. J Lightwave Technol 2012;30(8):1097–112. [15] Frank A, Bohnert K, Haroud K, Brändle H, Poulsen CV, Pedersen JE, et al. Distributed feedback fiber laser sensor for hydrostatic pressure. IEEE Photonic Technol Lett 2003;15(2):1758–60. [16] Hao Zhang, Jianhua Luo, Bo Liu, Shuangxia Wang, Chenglai Jia, Xiurong Ma. Polarimetric multilongitudinal-mode distributed Bragg reflector fiber laser sensor for strain measurement. Microw Opt Technol Lett 2009;51 (11):2559–63. [17] Durán-Sanchez M, Beltrán-Pérez G, Castillo-Mixcóatl J, Muñoz-Aguirre S, Méndez-Otero M. Experimental study of the fiber laser output intensity behavior and its application to a water flow sensor. Sensor Actuat B 2007;123:816–21. [18] Jianguan Tang, Junqiang Sun. Stable and widely tunable wavelength-spacing single longitudinal mode dual-wavelength erbium-doped fiber laser. Opt Fiber Technol 2010;16:299–303. [19] Álvarez Tamayo RI, Durán Sánchez M, Pottiez O, Kuzin EA, Ibarra Escamilla B. Tunable dual-wavelength fiber laser based on a polarization-maintaining fiber Bragg grating and a Hi–Bi fiber optical loop mirror. Laser Phys 2011;21 (11):1932–5.
Contents lists available at ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
Alternative interrogation method for a dual laser sensor based on fiber Bragg gratings to measure temperature using the fundamental beating frequency intensity O. Méndez-Zepeda, S. Muñoz-Aguirre n, G. Beltrán-Pérez, J. Castillo-Mixcóatl Facultad de Ciencias Físico-Matemáticas, Benemérita Universidad Autónoma de Puebla, 18 sur y Av. San Claudio, Col. San Manuel, CU, C.P. 72570 Puebla, México
art ic l e i nf o
a b s t r a c t
Article history: Received 20 August 2014 Received in revised form 11 October 2014 Accepted 16 October 2014 Available online 7 November 2014
In this work we present an alternative interrogation method for a dual sensor based on a fiber laser used to measure temperature in two remote locations simultaneously. The dual laser consisted of two Fabry– Perot cavities each conformed by two fiber Bragg gratings (FBG). For each cavity, one FBG was used as the reference and the other one as a sensing element. The sensing element interrogation was performed by the quantification of the fundamental beating frequency (FBF) intensity, which was calculated using the fast Fourier transform algorithm. The laser emissions were centered at 1549 and 1556 nm, while the lengths of cavities were of 300 and 400 m, which corresponds to FBFs of 334 and 258 kHz, respectively. The quantification of the temperature was calculated from the difference between the FBF values of both cavities. Such difference describes a geometrical plane in function of the two sensing FBGs temperatures. Consequently, it was possible to achieve temperature measurements in a range of 25–28 1C for the two sensors. & 2014 Elsevier Ltd. All rights reserved.
Keywords: Fiber Bragg grating Fundamental beating frequency Optical fiber laser sensor
1. Introduction In general, the sensors based on fiber Bragg gratings (FBG) require methods or techniques for measuring the spectral shifts that the Bragg wavelength (λB) presents when the FBG is in contact with some physical variable such as strain or temperature variations. Then, in order to detect and measure the changes of such variables through spectral shifts it is required expensive equipment such as the optical spectrum analyzer (OSA) [1], electrical spectrum analyzer (ESA) [2], tunable lasers [3], and different systems such as Sagnac filters [4,5], tunable Fabry–Perot filters [6], Mach–Zehnder interferometers [7,8], among others. Therefore, it is required to develop detection and measurement techniques that are economical and easy to implement. On the other hand, the measurement of two or more variables with the same setup using FBGs becomes difficult and expensive, although the development of some sensors that perform physical variables measurements through the quantification of FBGs reflection or transmission spectra shifts have been reported. For instance, some works use the time division multiplexing (TDM) [9], arrayed waveguide gratings (AWGs) demultiplexer [10], while
n
Corresponding author. Tel.: þ 52 222 2295500x2176; fax: þ52 222 2295636. E-mail addresses: [email protected], [email protected] (S. Muñoz-Aguirre). http://dx.doi.org/10.1016/j.optlastec.2014.10.014 0030-3992/& 2014 Elsevier Ltd. All rights reserved.
another employs the wavelength-division multiplexing (WDM) technique [11] or in the case of Lu et al. that report a simultaneous discrimination of axial strain and temperature FBGs through the wavelength shift [12]. In general, all the works above mentioned use complicated experimental setups. Other laser sensors based on FBGs cavities involve the use of the determination of the fundamental beating frequency (FBF) shifts to quantify the physical variables. For instance, a sensor principle based on the measurement of the beating frequency shifts, which are provoked by birefringence changes, has been reported. In this category we can find the polarimetric fiber laser sensors, which require two orthogonal polarization modes to generate polarization mode beating as the sensing signal [13–16]. They focus mainly in the determination of the FBF shifts. In a previous work we have reported that there exists a relationship between the output laser intensity and the overlapping region of two FBGs reflection spectra caused by one FBG stretching as result of a physical variable [17]. It was also found that the output optical power variations were linear in relation to the sensing FBG stretching. However, if this setup is used to evaluate more than one variable simultaneously, it would not be possible to discriminate the variables through the optical power intensity, since the total response is the combination of both intensities; if there is an intensity variation it would not be possible to know which cavity is the one that provokes such variation. The cavity discrimination would be possible if the output optical power of the dual laser were analyzed in the frequency domain, in such case,
160
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
the response of each sensor would be identified from the FBF corresponding to each cavity. The last affirmation is based on the fact that the length of a laser cavity determines the beating frequencies between any pair of resonance modes inside the cavity that can be calculated from Equation (1).
νq ¼ q
c 2d
q ¼ 1; 2; 3; …
ð1Þ
where c is the light speed inside the cavity and d is the cavity length. For the particular case when q¼ 1 (beating between two adjacent modes), the FBF is given by Equation (2). FBF ¼ Δν ¼
c 2d
ð2Þ
From here, if two cavities with different length are used, it can be expected that each laser emission can be identified by their FBF value according to Equation (2). On the other hand, if the dual laser output is analyzed in the frequency domain, each FBF intensity change as a function of the laser power intensity which depends on the FBGs spectra overlapping. It is important to mention that the quantification and identification using the FBF intensity has not been used so far and that its implementation would decrease cost and make the interrogation setup simpler, especially when the system is used to measure two variables simultaneously. In this work we propose an alternative simple method to interrogate a dual temperature sensor, where each sensor was set in a different location. The interrogation was performed by the determination of the FBF intensities of a multilongitudinal dual laser, which has a few advantages for sensing applications over the above mentioned schemes. First, the laser cavities were constructed with commercial elements, avoiding the use of special FBG written in the erbium doped fiber (EDF) and the polarization states control that are quite sensitive to fiber bends and twists, which must be avoided especially when the physical variable produces small changes in the birefringence. Secondly, a large cavity provides much higher laser power than single-mode lasers, and hence a higher SNR can be obtained. Furthermore, our scheme can be used as a dual and remote sensor since the standard fiber that conform part of the cavity can be installed far away from the system. On the other hand, the FBF can be measured with a conventional data acquisition (DAQ) card, without using expensive equipment such as the optical or electrical spectrum analyzers (OSA and ESA, respectively).
2. Experimental The experimental setup is shown in Fig. 1. The dual fiber laser consisted of two Fabry–Perot cavities with lengths of 300 and 400 m, respectively, constructed with FBGs and standard optical fiber SMF28 (Corning). Both cavities share 7 m length of EDF as the gain medium.
These lengths are important since the fiber can be installed far away from the interrogation system and it would allow using the sensors in two remote locations. Furthermore, the FBF values of long cavities are in the order of hundreds of kHz, which is not so high frequency and it can be measured using conventional electronic circuits. The gain medium was pumped by a laser diode (26-8052-100, JDSU) with a wavelength of 980 nm, which was adjusted to an output optical power of 35.5 mW and was coupled to the EDF through a wavelength division multiplexer (WDM). Each cavity was constructed with two FBGs (Bragg Photonics Inc.), one of them was the reference element (FBG-R) and the other one was the sensing element (FBG-S). The two reference elements were FBG-R1 (λB ¼1549 nm, reflectivity of 58%) and FBG-R2 (λB ¼1556 nm, reflectivity of 58%), while the sensing elements were FBG-S1 (λB ¼ 1546 nm, reflectivity of 58%) and FBG-S2 (λB ¼1553 nm, reflectivity of 59%). Without any excitation all the FBGs λB are different and there is not laser emission since there is not conformed any cavity. However, when the spectrum of the FBG-S starts overlapping over that of FBG-R, provoked by a physical variable, the cavity is set and the laser intensity starts and it changes as a function of the spectral overlapping of the FBGs in each cavity. Moreover, if we try to measure temperature with this FBGs, we need a very high temperature variations, since the spectral shifts of the FBGs are quite slight (approximately 10 pm/1C). For this reason, the system needs an initial condition of spectra overlapping which results in an initial dual emission that is reached when we apply a stretching on the sensing FBG through a micrometer screw, which is not shown in the scheme of Fig. 1 since it was used only to set such initial condition. It is important to mention that this initial stretching does not affect the FBF, since it is a stretching of a few tens of micrometers in hundreds of meters of the total cavity length. After reaching this initial condition, we can start to measure temperature variations. The temperature of each FBG-S was controlled by a Peltier cell and only a portion of 4 cm of the fiber that contains the FBG was exposed to temperature changes. This was performed for each cavity. The signal conditioning electronic circuit was composed of a conventional photodiode (FGA10, detection range from 800 to 1800 nm) and a transimpedance amplifier, which provides an output voltage in the range from 1.2 to 4.8 V dc and a bandwidth of 500 kHz. These values were adequate for the frequency range that we were measuring in our system, whose maximum frequency was around 334 kHz. Finally, the obtained signal was digitalized using a DAQ card (PCI-1712, ADVANTECH) with a sampling rate of 1 MS/s and a 12-bit resolution. This signal was analyzed using the FFT algorithm to determine its spectrum in the frequency domain and then to determine the frequency value and amplitude of the FBF. The data acquisition procedure was as follows: a measurement of 222 samples (during approximately 5 s) was performed and the obtained data vector was segmented in sub-vectors of 28, since the FFT algorithm works better with data sets of 2n sizes. Then, the FFT was calculated in order to determine the FBF intensity for each segment. Finally, the behavior of the FBF intensity in function of the FBG-S temperature was studied.
Fig. 1. Experimental setup.
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
FBF intensity x 10-3 (a.u.)
10 FBF = 328 kHz 8 6 4 2 0 0
100
200 300 Frequency (kHz)
400
500
Fig. 2. FFT of the acquired signal for a single cavity.
1.0
FBF intensity x 10-3 (a.u.)
R2 = 0.9791 0.8 0.6 0.4 0.2 25
26 27 FBG-S temperature (°C)
28
Fig. 3. FBF intensity as a function of the FBG-S temperature for a single cavity.
3. Results and discussion First of all, the study of only one cavity was performed to measure and determine the FBF. Therefore, it was carried out a measurement of the sensor output signal during 256 μs (28 data) at a rate of 1 MS/s and the FFT algorithm was applied to these data. In Fig. 2, the spectrum in the frequency domain of the single sensor output signal is shown. A peak located in 328 kHz, which corresponds to the FBF for a 314 m cavity length is clearly observed. Such FBF value is according with the one calculated from Equation (2), which was 324 kHz. The FBF measurement resolution was of 3.9 kHz. As it was already mentioned above, it can be expected that the intensity of the FBF peak changes as a function of the spectra overlapping of the FBGs, which is originated by the temperature variations detected by the FBG-S. In order to corroborate this statement, the FBF intensity changes in relation to the temperature in a range of 25–28 1C in steps of 1 1C were performed for one cavity configuration. The result is shown in Fig. 3. The FBF intensity has a slight offset; since an initial emission was set in a suitable point where both FBG spectra were slightly overlapped. The temperature measurement was performed from this point. The FBF intensity has a linear tendency in function of temperature with a correlation factor of R2 ¼ 0.9791, which indicates a quite good correlation when a linear fitting was performed. This behavior is similar to the case when the relationship between optical power and the FBG spectra overlapping was studied (inset in Fig. 3) and it is in agreement with results obtained in a previous work [17]. This fact indicates that the FBF intensity has a direct relationship with the spectra overlapping area of both FBG that conform the cavity. For this reason, we can say that the alternative
161
method proposed in this work can be used as a measurement method of a physical variable that change the spectra overlapping of the FBG, which can be stretching or temperature changes. In this particular case, it was possible to measure temperature variations with sensitivity of 1.68 10 4 1C 1, and a maximum error of 6%, when various measurements were performed, as is shown by the error bars in Fig. 3. Such sensitivity depends on the spectra width and profile of the FBGs that compose the cavity. It is important to mention that the cavities are conformed by FBGs that has a spectral width of 1.2 nm, and due to the initial spectral overlapping we have a quite short range to measure temperature. However we are interested in showing an alternative interrogation method to measure two variables based on the quantification of the FBF intensity instead of FBF shifts, which requires expensive measurement equipment. The measurement range could be increased by changing the FBGs with wider reflection spectra, although a further study would be necessary. For the purpose of measuring two variables simultaneously (temperature in two remote locations), it is important to obtain a dual emission, which can be achieved with two cavities in the same setup (Fig. 1). As it was already mentioned above, it is not possible to identify which cavity is emitting or its emission intensity when the measurement is performed using just one photodiode to detect optical power changes. With the method proposed in this work, each emission can be identified in the frequency domain employing the corresponding FBF value for each cavity. It is important to remark that it is quite difficult to obtain a stable and simultaneous dual emission [18,19]. In our case, we could achieve such simultaneous and stable dual emission, whose peaks can be controlled in an independent way. An example of the dual emission measured with an OSA is shown in Fig. 4a. The two peaks can be clearly observed, whose wavelengths are centered at the λB of the corresponding FBG-R; i.e. at 1549 and 1556 nm, respectively. On the other hand, the corresponding FBF value for each emission was determined and it was found that those values were in agreement with the cavities lengths. In Fig. 4b, the FBF spectrum of both cavities is shown, where two peaks can be clearly observed centered at 258 y 334 kHz. From now on we will call them FBF1 and FBF2, respectively. According to Eq. (2), the frequency peaks correspond to cavity lengths of 400 and 300 m, respectively. With this, we show that it is possible to identify, in an independent way, the emissions of the cavities and analyze the behavior of the FBF intensities when the two FBG-S's are under the influence of a physical variable. In this case, we can measure temperature changes by using just one conventional photodiode. Once the dual emission was achieved, the behavior of the FBF1 and FBF2 intensities in function of the temperature changes of both FBG-S (FBG-S1 and FBG-S2) was analyzed. The measurement of the FBF intensity changes of both cavities was performed under the following conditions: First, the FBG-S1 was kept at a constant temperature while that of the FBG-S2 was swept from 25 to 28 1C in steps of 1 1C. This process was repeated for FBG-S1 temperatures of 26, 27 and 28 1C, respectively. The results of the measurement of both FBFs intensities for these conditions are shown in Fig. 5. For each case, we can observe an increment of the FBF2 intensity as the FBG-S2 temperature was increased. This increment is in agreement with the results shown in Fig. 3. On the other hand, one would expect that the FBF1 intensity remain constant for each FBG-S2 temperature sweep, however a decrement was observed. Such decrement is more evident for the case when FBG-S1 was at 28 1C (Fig. 5d). This could be due to the modes competition inside the cavities since both of them share the same gain medium and probably the gain increment of the cavity 2 would affect the gain of the cavity 1 and vice versa, as can be observed in Fig. 5, a and d. The same behavior was observed in the opposite case, that is, when the FBG-S2 was under a constant temperature while the FBG-S1
162
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
0
Laser emission1
Laser emission 2
1549 nm
1556 nm
10 FBF1 258 kHz
(a.u.) -3
-20
FBF intensity x 10
Optical power (dBm)
-10
-30 -40 -50
8 6 FBF2 334 kHz
4 2 0
1545
1550
1555
0
1560
100
200
300
400
500
Frequency (kHz)
Wavelength (nm)
Fig. 4. (a) Emission of the two cavities, centered at 1549 and 1556 nm, (b) FBF intensity corresponding to each emission in the frequency domain.
4 2 0
8
FBF1 FBF2
6 4 2 0
10 8 6 4 2 0
26 25 26 26 26 27 26 28 Temperature (°C)
25 25 25 26 25 27 25 28 Temperature (°C)
FBF1 FBF2
10
FBF intensity x10-3 (a.u)
6
10
FBF intensity x10-3 (a.u)
8
FBF1 FBF2
FBF intensity x10 -3 (a.u)
FBF intensity x10-3 (a.u.)
10
FBF1 FBF2
8 6 4 2 0
27 25 27 26 27 27 27 28 Temperature (°C)
28 25 28 26 28 27 28 28 Temperature (°C)
Fig. 5. FBF intensity changes as a function of temperature for the two cavities.
4 2 0 25 25 26 25 27 25 28 25 Temperature (°C)
8
FBF1 FBF2
6 4 2 0 25 26 26 26 27 26 28 26 Temperature (°C)
10 8
FBF1 FBF2
6 4 2 0 25 27 26 27 27 27 28 27 Temperature (°C)
10
FBF intensity x10-3 (a.u.)
6
10
FBF intensity x10-3 (a.u.)
8
FBF1 FBF2
FBF intensity x10 -3 (a.u.)
FBF intensity x10-3 (a.u.)
10
FBF1 FBF2
8 6 4 2 0
25 28 26 28 27 28 28 28 Temperature (°C)
Fig. 6. FBF intensity changes as a function of temperature for the two cavities.
temperature was swept in the range from 25 to 28 1C, as it is shown in Fig. 6. Another important aspect is that if we focus only in one of the FBFs intensities, for instance the FBF2, we can observe that the slope decreases as the FBG-S1 temperature increases. Such effect can be used to have a sensor with variable sensitivity in FBG-S2, which is controlled by the temperature of the FBG-S1. According to the above results, since both FBF intensities changed at the same time, it would be possible to find a relationship of the difference between both intensities (DFBF ¼FBF2 FBF1). For this reason, it was decided to characterize such behavior by calculating the differences of the FBF intensities for each pair of temperature values (T1 and T2) of the FBG-S1 and FBG-S2, respectively. The obtained results are shown in Fig. 7, where it can be observed that the DFBF describes a plane, which has an inclination in the direction of both independent axis, that correspond to the temperature of both FBG-S's T1 and T2. Therefore, it is possible to determine a pair of
temperature values for each DFBF. The equation that describes this plane was calculated from a fitting of the experimental measurements (Eq. 3), which had a correlation coefficient of 0.9830. Such correlation value is adequate to determine T1 and T2. DFBF ¼ –1:44 10 3 T1 þ 2:03 10 3 T2–13:84 10 3
ð3Þ
From Equation (3) is possible to quantify the temperature of each FBG-S, simultaneously. The temperature evaluation is as follows: First, the DFBF is calculated from the measured values FBF1 and FBF2. Then, the plane equation is evaluated by sweeping with slight increments all the possible temperature values in the range of 25– 28 1C for both FBG-S´s. At this point, when the calculated FBF difference equals the measured value, we can say that the temperature of each FBG-S's has been determined. Finally, from Equation (3), it is important to mention that the sensitivity depends on the temperature of both FBG-S. Since there are two sensors, two
O. Méndez-Zepeda et al. / Optics & Laser Technology 67 (2015) 159–163
163
Acknowledgments The present work was partially supported by the projects CONACYT Jóvenes Investigadores 61126, VIEP-BUAP CAMJ-EXC14I, BEPG-EXC14-I, and MUAS-EXC14-G.
References
Fig. 7. DFBF of the measured data and fitting of the experimental measurements (shadowed plane).
sensitivities defined by the temperature of each FBG-S, were obtained. In this case, we obtained a sensitivity of 1.44 10 3 1C 1 for the sensor 1 and 2.03 10 3 1C 1 for the sensor 2. It is clear that this sensitivity is defined for the DFBF. The temperature range that can be measured depends on the reflection spectra width of the FBGs, in our particular case, before starting the temperature measurement, it was applied an initial strain to the FBG-S with the purpose of adjusting the laser intensity in a value where it were possible to achieve the dual emission and to determine the FBF of both emissions, for this reason the dynamic range was limited to the above mentioned values. Such dynamic range probably can be extended if FBGs with wider reflection spectra were used, although a further study would be necessary to determine if both FBGs that compose each cavity must be wide or only one of them. This study is out of the scope of the present paper, since we are only interested in showing the concept of the dual sensor with two cavities working simultaneously. 4. Conclusions In this work, we presented an alternative method to measure temperature in two different locations simultaneously using a dual laser. It consisted of two Fabry–Perot cavities composed by FBGs that share a single gain medium. The setup has the advantage that it is quite cheap and employs conventional elements. The FBF intensities of the dual laser were determined by a data processing performed with a commercial DAQ and a computer program that uses the FFT algorithm. The results showed that the FBF intensity was proportional to the spectra overlapping of the FBGs that compose each cavity. Moreover, when two cavities were used, it was possible to measure temperature changes in two remote locations in the range of 25–28 1C by evaluating the difference between both FBF intensities (DFBF). It was found that DFBF defines a geometrical plane in function of the temperature of both FBG-S's. Finally, although DFBF depends on both FBF signals, the two sensors can be operated in an independent way with their own sensitivity, which in this case, was constant for the temperature of each FBG-S.
[1] Yasukazu Sano, Toshihiko Yoshino. Fast optical wavelength interrogator employing arrayed waveguide grating for distributed fiber Brag grating sensors. J Lightwave Technol 2003;21(1):132–9. [2] Shengchun Liu, Zuowei Yin, Liang Zhang, Liang Gao, Chen Xiangfei, Cheng Jianchun. Multilongitudinal mode fiber laser for strain measurement. Opt Lett 2010;35(6):835–7. [3] Chan CC, Gong JM, Jin W, Demokan MS. Investigation of unwanted interferometric signals in a fiber Bragg grating sensor using a tunable laser and a first derivative interrogation technique. Opt Commun 2000;173:203–10. [4] Xiufeng Yang, Chun-Liu Zhao, Qizhen Peng, Zhou Xiaoqun, Chao Lu. FBG sensor interrogation with high temperature insensitivity by using a HiBi-PCF Sagnac loop filter. Opt Commun 2005;250:63–8. [5] Kyung-Rak Sohn. Fiber Bragg grating-tuned feedback laser flow sensor system. Sensor Actuat A 2012;179:1–4. [6] Kersey D, Berkoff TA, Morey WW. Multiplexed fiber Bragg grating strainsensor system with a fiber Fabry–Perot wavelength filter. Opt Lett 1993;18 (16):1370–2. [7] Ya-nan Zhang, Yong Zhao, Wang Qi. Improved design of slow light interferometer and its application in FBG displacement sensor. Sensor Actuat A 2014;214:168–74. [8] Qiqi Yao, Hongyun Meng, Wei Wang, Hongchao Xue, Xiong Rui, Huang Ben, et al. Simultaneous measurement of refractive index and temperature based on a core-offset Mach–Zehnder interferometer combined with a fiber Bragg grating. Sensor Actuat A 2014;209:73–7. [9] Sun Q, Liu D, Wang J, Liu H, Xia L, Shum P. Experimental demonstration of multi-point temperature warning sensor using a multi-channel matched fiber Bragg grating. IEEE Photonic Technol Lett 2008;20(11):933–5. [10] Hongqiang Lia, Yang Lia, Enbang Lib, Xiaye Donga, Yaoting Baia, Yu Liua, et al. Temperature-insensitive arrayed waveguide grating demodulation technique for fiber Bragg grating sensors. Opt Laser Technol 2013;51:77–81. [11] Chang-Seok Kim, Tae Ho Lee, Yun Sik Yu, Young-Geun Han, Sang Bae Lee, Myung Yung Jeong. Multi-point interrogation of FBG sensors using cascaded flexible wavelength-division Sagnac loop filters. Opt Express 2006;14 (19):8546–51. [12] Lu P, Men L, Chen Q. Resolving cross sensitivity of fiber Bragg grating with different polymeric coatings. Appl Phys Lett 2008;92:171112. [13] Kringlebotn JT, Loh WH, Laming RI. Polarimetric Er3 þ -doped fiber distributed-feedback laser sensor for differential pressure and force measurements. Opt Lett 1996;21(22):1869–71. [14] Bai-Ou Guan, Long Jin, Yang Zhang, Hwa-Yaw Tam. Polarimetric heterodyning fiber grating laser sensors. J Lightwave Technol 2012;30(8):1097–112. [15] Frank A, Bohnert K, Haroud K, Brändle H, Poulsen CV, Pedersen JE, et al. Distributed feedback fiber laser sensor for hydrostatic pressure. IEEE Photonic Technol Lett 2003;15(2):1758–60. [16] Hao Zhang, Jianhua Luo, Bo Liu, Shuangxia Wang, Chenglai Jia, Xiurong Ma. Polarimetric multilongitudinal-mode distributed Bragg reflector fiber laser sensor for strain measurement. Microw Opt Technol Lett 2009;51 (11):2559–63. [17] Durán-Sanchez M, Beltrán-Pérez G, Castillo-Mixcóatl J, Muñoz-Aguirre S, Méndez-Otero M. Experimental study of the fiber laser output intensity behavior and its application to a water flow sensor. Sensor Actuat B 2007;123:816–21. [18] Jianguan Tang, Junqiang Sun. Stable and widely tunable wavelength-spacing single longitudinal mode dual-wavelength erbium-doped fiber laser. Opt Fiber Technol 2010;16:299–303. [19] Álvarez Tamayo RI, Durán Sánchez M, Pottiez O, Kuzin EA, Ibarra Escamilla B. Tunable dual-wavelength fiber laser based on a polarization-maintaining fiber Bragg grating and a Hi–Bi fiber optical loop mirror. Laser Phys 2011;21 (11):1932–5.