We propose and experimentally demonstrate monitoring of a fiber fuse in real time using an optical time domain reflectometer (OTDR). When a fuse starts, a weak reflection of light occurs from the leading edge of the fuse where plasma and voids are being formed in the core. In this work, we examined the possibility of monitoring a fiber fuse from a remote location using an OTDR. We demonstrate a method that allows us detect a fuse progressing at remote locations (over kilometers away). It was found to be effective even in the presence of strong spurious backscattering, such as spontaneous Raman scattering due to a strong continuous wave pump. Moreover, from the progress of the reflection edge monitored by the OTDR, the fuse velocity could be readily determined.
©2010 Optical Society of America
Optical fibers carrying high power laser radiation in excess of few watts are prone to damage caused by a fiber fuse, which may be caused by localized heating of the core due to dust accumulated at the connector or fiber end face, as well as leakage at tight bends [1–5]. Once it occurs, the fiber fuse continues to propagate as long as the power remains above a certain power level, known as fuse threshold power. Fuse threshold is only about 1.3-1.5 watt for single mode fiber [6–9], which is the upper limit for safe transmission of laser radiation in single mode fibers. For transmitting light at higher power levels, the chance that a fuse may occur increases, and hence it is crucial to develop techniques to protect the source and fibers from this catastrophic damage. A short tapered fiber section with a large mode field diameter [8,10] or special hole-assisted fiber  can be incorporated in the fiberoptic path to terminate a fuse. However the ability to quickly detect a propagating fuse remotely would give the opportunity to cut off the optical power, minimize the damage to the fiber, and protect the laser system.
We have recently observed that during propagation of a fuse, the region of plasma and voids in the leading edge can cause a minute reflection of light that experiences amplitude modulation at certain characteristic frequencies . By monitoring of the RF spectrum of the backreflected light, we have also developed a fiber fuse detection scheme [11–13]. However one shortcoming of this scheme is that it is unable to determine the exact location of the fuse, which would be necessary for repairing or replacement of the damaged fiber.
In this paper, we demonstrate monitoring of a propagating fiber fuse in real time using an OTDR. Backreflection of light which occurs during fuse propagation is detected remotely from a kilometer away using an OTDR. The monitoring scheme could detect fuse even in the presence of strong spurious backscattering, such as spontaneous Raman scattering due to the strong pump. Moreover, we show a simple way to measure the speed of the fuse propagation from the evolution of the reflection edge monitored by the OTDR.
Earlier, a monitoring and testing system for maintenance of passive optical networks (PONs)  using an OTDR has been reported. However, there is a major difference between the nature of the fault in PONs and that of a fiber fuse. In PONs, faults remain stationary, i.e. the reflection edge observed in OTDR is time-invariant. On the contrary, when a fuse occurs, the reflection point shifts as the fuse moves towards the source end. As we will show in the following, the integration or averaging that is commonly performed to enhance the sensitivity of OTDRs can readily be used to extract information on the speed of fuse.
The experimental setups that we used for real time monitoring of a fiber fuse using an OTDR are shown in Figs. 1(a) and 1(b). We considered two distinct cases; (1) The OTDR operated at a wavelength close to wavelength of the high power laser source. (2) The OTDR operated at a longer wavelength which falls within the Raman gain spectrum induced by the high power laser source. In the later case, the weak backreflected OTDR signal might sink in the spontaneous Raman scattering.
In the experimental setup shown in Fig. 1(a), an external cavity semiconductor laser operating at a wavelength of 1560 nm, was amplified to about 36 dBm and applied into a ~25-meter long SMF-28 fiber through a 10 dB optical coupler. We used a commercial OTDR (ANDO) that was connected through the 10% port of the fiber coupler. The OTDR interrogating pulses were 10-ns-long, and had a peak wavelength of ~1540 nm and a spectral width of about 15 nm. A tunable bandpass filter (FWHM bandwidth of 15 nm) was also incorporated, which allowed OTDR signal to transmit and prevented backreflected pump light from reaching the instrument. In addition a long piece of SMF fiber (~650 m) was also installed in the path of the OTDR to remain outside the dead zone of OTDR measurement.
In Fig. 1(b), a high power cascaded Raman laser operating at a wavelength of 1480 nm was used to initiate a fuse, while an OTDR was operated at a wavelength of 1540 nm. The laser and the OTDR were connected through a 1480/1550 WDM coupler. While the WDM coupler prevented the pump from reaching the OTDR, a bandpass filter was also placed after the OTDR to prevent the broad Raman ASE induced by the pump from entering the OTDR.
We spliced a few cm long DCF (dispersion compensating fiber) at the distal end of the fiber and initiated fuse by applying light absorbing white paint (correction fluid used as stationery) at the fiber end while emitting about 2.5 W. DCF fiber has a lower threshold for fuse, thus making it easier to initiate. OTDR traces were recorded with an integration time of 30 seconds.
3. Results and discussions
3.1 Wavelengths of high power laser and OTDR are near 1.55 μm
When a fuse occurs, the Fresnel reflection comes mostly from the leading edge of the fuse, as the damaged part of the fiber exhibits high loss (~2.6 dB/cm) . Consequently, as the fuse was started, we instantaneously observed a sharp rise in the OTDR trace that corresponded to the leading edge of the fuse. Moreover, as the fuse propagated, and OTDR measurements were integrated over a fixed time interval, and we observed a characteristic flat region in the trace. Figure 2(a) shows a trace measured with an integration time of 30 sec. The integration time represents the time during which traces were recorded and averaging was performed. The launched power was kept constant at 2.45 W. The width ΔL of this flat plateau was found to be proportional to the integration time, as the leading edge of the fuse where reflection occurred advanced with time. From the width of this plateau ΔL and the time over which measurement was performed, Δt, we could readily measure the speed of fuse propagation, i.e. v = ΔL/Δt. From Fig. 2(a) we obtained ΔL of 8.5 m and a corresponding Δt of 19.6 sec, we obtained a fuse velocity of 0.43 m/s.
We also took OTDR measurements after the fuse was terminated. Figures 2(b) and 2(c) are post-fuse OTDR traces taken for two different lengths of damaged fiber, 10m and 24 cm, respectively. For both measurements, the traces were similar, and showed narrow reflection peaks, which was in contrast to the trace measured in the presence of fuse [Fig. 2(a)]. In a recent study, we have found that the light attenuates severely in the fiber damaged by fuse and measured an attenuation loss of about 2.6 dB/cm. Therefore reflection of light occurs from a damaged fiber is localized from the leading edge of the fused fiber, which is shown in Figs. 2(b) and 2(c). The steep decay of the reflection value on the right side of the peak is due to artifacts of measurement resolution.
3.2 Wavelength of OTDR overlapping with the Raman gain spectrum of the pump
We chose a pump source with at a shorter wavelength, to study the effect of pump induced spontaneous Raman scattering on the OTDR measurement. We found that the noise could be greatly reduced by placing a bandpass filter following the OTDR. Figures 3(a) -3(d) show the OTDR traces measured with the filter in under the four cases; (a) pump off, (b) pump on, (c) during fuse and (d) after terminating the fuse.
Similar experiment was also performed after removing the bandpass filter, and the results are shown in Figs. 3(e)–3(h). Similar to previous experiment, we can clearly see the plateau in the OTDR trace shown in Fig. 3(g). However, in Figs. 3(f), 3(g), we can see an increase in the background at location beyond the front edge of the fuse, which is due to the SRS entering the instrument. The use of a bandpass filter can therefore significantly reduce the unwanted noise level and improve the detection sensitivity.
From the width of this plateau ΔL and the time Δt over which integration was performed, we could readily measured the speed of fuse propagation using v = ΔL/Δt. From Fig. 3(c) we obtained ΔL of ~14 m and a corresponding Δt of 30 s, we obtained a fuse velocity of 0.46 m/s.
We also performed OTDR measurements with different integration times, after the fuse was initiated. Figure 4 shows OTDR traces taken for three different times of intergration; that are 10 s, 20 s and 30 s. These plots clearly show that the width of the plateau increases proportionately with the integration time, allowing us determine the fuse velocity using, using v = ΔL/Δt.
Based on the proposed scheme, one can build a fuse protection system, where the backreflection from a transmission line carrying high power will be monitored continuously using an OTDR, and comparison of the successive traces will be made. A fuse can be identified by any shift (towards the source) of a reflection peak seen in the successive traces. A simple modification of the software processing unit of commercial OTDRs should be able to perform this task as well as to provide an alarm signal when such event occurs, which will be useful to turn off the optical source using its interlock feature.
3.3 Velocity and threshold of fuse propagation
From the experimental results shown in the previous section, it can be clearly seen that an OTDR can be used to effectively detect propagation of a fiber fuse and measure its velocity. To compare the results of OTDR measurement, fuse velocity was also measured directly through visual observation of fuse propagation.
Figure 5 represents the fuse velocity measured for different pump power levels for two different pump wavelengths. The measured fuse velocity for a 1560 nm pump with a launched power of 2.45 W was 0.42 m/s, which was in excellent agreement with that obtained from the OTDR measurement. Similarly, for experiment using a pump at 1480 nm, we obtained a fuse velocity of 0.46 m/s (for a launched pump power of 2.35 W), which agrees well with the speed measured using OTDR.
The fuse threshold powers for pumping at these wavelengths were also measured by lowering the pump power to a level that halts fuse propagation. The fuse threshold at 1560 nm and 1480 nm was found to be 1 0.336 W and 1.39 W, respectively. This indicated that when the wavelength was decreased by 5.12% (from 1.56 μm to 1.48 μm), we observed a decrease in fuse threshold by 3.88%.
It has been earlier reported [8,9] that the threshold power vary linearly with the mode-field radius ωo, which is related to core radius a and the V parameter by the relation, ωo ≈a(0.65 + 1.619/V1.5 + 2.879/V6) = a.F(V) . Here, V = 2πa(NA)/λ and F(V) denotes a function of V enclosed by the bracket. Any change in the pump wavelength (dλ) will result in a change in the threshold power (dPTh), in a way that can be expressed by,
Figure 6 plots the ratio of fractional change in fuse threshold and fractional change in the pump wavelength, (dPT h/PTh)/(dλ/λ) as a function of V parameter. Using Δλ/λ = −0.0512 and V = ~2.2, we estimate a decrease in fuse threshold by −0.039, that agrees well with that we found experimentally.
In summary, we propose and experimentally demonstrate real time monitoring of a fiber fuse performed using an OTDR. We could easily identify a moving fuse in remote locations and accurately measure the velocity of fuse propagation. The proposed system can be easily incorporated in systems carrying high power, such as for pumping of remote EDFA or systems using distributed Raman amplification and the measurement remain unaffected by the nature of the pump source.
References and links
1. R. Kashyap and K. J. Blow, “Observation of catastrophic self-propelled self-focusing in optical fibers,” Electron. Lett. 24(1), 47–49 (1988). [CrossRef]
3. Y. Shuto, S. Yanagi, S. Asakawa, M. Kobayashi, and R. Nagase, “Fiber fuse generation in single-mode fiber-optic connectors,” IEEE Photon. Technol. Lett. 16(1), 174–176 (2004). [CrossRef]
4. R. M. Percival, E. S. R. Sikora, and R. Wyatt, “Catastrophic damage and accelerated aging in bent fibers caused by high optical powers,” Electron. Lett. 36(5), 414–416 (2000). [CrossRef]
5. E. M. Dianov, I. A. Bufetov, A. A. Frolov, Y. K. Chamorovsky, G. A. Ivanov, and I. L. Vorobjev, “Fiber fuse effect in microstructured fibers,” IEEE Photon. Technol. Lett. 16(1), 180–181 (2004). [CrossRef]
6. Y. Shuto, S. Yanagi, S. Asakawa, M. Kobayashi, and R. Nagase, “Fiber fuse phenomenon in step-index single-mode optical fibers,” IEEE J. Quantum Electron. 40(8), 1113–1121 (2004). [CrossRef]
8. K. Seo, N. Nishimura, M. Shiino, R. Yuguchi, and H. Sasaki, “Evaluation of High-power Endurance in Optical Fiber Links,” Furukawa Rev. 24, 17–22 (2003).
9. K. Takenaga, S. Omori, R. Goto, S. Tanigawa, S. Matsuo, and K. Himeno, “Evaluation of high-power endurance of bend-insensitive fibers,” in Optical Fiber Communication Conference, Technical Digest (Optical Society of America, 2008), paper JWA11.
10. D. P. Hand and T. A. Birks, “Single mode tapers as fiber fuse damage circuit breakers,” Electron. Lett. 25(1), 33–34 (1989). [CrossRef]
12. K. S. Abedin, and T. Morioka, “Remote detection of fiber fuse propagation in optical fibers,” in Optical Fiber Communication Conference, Technical Digest (Optical Society of America, 2009), paper OThD5.
13. K. S. Abedin, T. Miyazaki, and M. Nakazawa, “Measurements of spectral broadening and Doppler shift of backreflections from a fiber fuse using heterodyne detection,” Opt. Lett. 34(20), 3157–3159 (2009). [CrossRef] [PubMed]
14. N. Tomita, H. Takasugi, N. Atobe, I. Nakamura, F. Takaesu, and S. Takashima, “Design and Performance of a Novel Automatic Fiber Line Testing System With OTDR for Optical Subscriber Loops,” J. Lightwave Technol. 12(5), 717–726 (1994). [CrossRef]
15. A. Ghatak, and K. Thyagarajan, Introduction to fiber optics, Cambridge University Press, NewYork, 1998.