Optics: measuring and testing – For optical fiber or waveguide inspection
Reexamination Certificate
1999-06-28
2003-04-01
Font, Frank G. (Department: 2877)
Optics: measuring and testing
For optical fiber or waveguide inspection
C385S012000
Reexamination Certificate
active
06542228
ABSTRACT:
The present invention relates to an optical time domain reflectometry (OTDR) method and apparatus.
Optical time domain reflectometry (see Barnoski, M. K. and Jensen, S. M., Applied Optics 1976, vol. 15, pp 2112-15) involves launching a short pulse of light into an optical fibre and observing the backscatter return from the entire length of the fibre. The backscatter consists of light scattered through a variety of mechanisms, including Rayleigh, Brillouin and Raman scattering. The scattered light is quasi-isotropic and that fraction of the light which falls within the cone of acceptance of the fibre in the reverse direction is guided back towards the source. The light signal thus obtained typically takes the shape of a decaying waveform, the rate of decay being indicative of the local attenuation of the fibre. However, in addition to changes in rate of decay, localised changes in signal level can be caused by localised variations of the scattering coefficient, of the numerical aperture (for multimode fibres), or of the spot size (for single mode fibres). The distance along the fibre can be related to time of arrival of the signal by means of the known velocity of light (in a manner similar to that used in other reflectometric techniques, such as Radar or Sonar).
In the case of telecommunications applications, where OTDR is most widely used, the interest is in determining the attenuation of the fibre as a function of distance and any changes in the loss with time or position (e.g. point discontinuities).
Some of the localised effects could be caused by the action of external measurands and this fact has been exploited in a variety of designs of distributed sensor (see Hartog, A. H., J. Lightwave Technology, 1983, vol. LT-1, pp 498-509). In those designs which have been developed commercially, a small part of the scattered light spectrum, consisting of Raman or Brillouin scattered light, is selected. These spectral lines are typically very weak compared with the dominant Rayleigh scattering, and a major problem in the design of such sensors is achieving a sufficient signal-to-noise ratio to obtain a measurement of adequate resolution in an acceptable measurement time.
In the cases of both OTDR and of distributed sensing using OTDR, one major limitation is that of the power which can be launched into the fibre. The performance of optical time domain reflectometers (OTDRs) and OTDR-based sensors is measured by the maximum length of fibre which can be measured to a given signal uncertainty in a given measurement time with a given spatial resolution. The length of fibre is itself determined by the ratio of the dynamic range of the instrument to the loss per unit length of the fibre measured. Since the losses vary between fibres, a more common description of the range of an OTDR is the dynamic range, i.e. the maximum one-way signal attenuation at which the backscatter signal(s) can be measured to the required resolution.
The dynamic range is determined principally by the energy of the probe pulse launched into the fibre, the sensitivity of the receiver and, although these cannot always be controlled by the instrument designer, by the characteristics of the fibre and the efficiency of the optical arrangement within the instrument. Thus the range of an OTDR or OTDR-based sensor is maximised by making the receiver as sensitive as possible and launching as much energy as possible into the fibre. The energy of the pulse may be increased by increasing either its peak power or its duration. In the latter case, the spatial resolution of the instrument (i.e. its ability to distinguish separate, but closely adjacent, features along the fibre) is degraded. The central problem is thus one of increasing the energy launched into the fibre without degrading the spatial resolution.
Whereas the technology of semiconductor lasers until recently limited the power available within optical fibres, especially single-mode fibres, to approximately 100 mW, the development in recent years of optical amplifiers, especially those based on rare-earth-doped fibres, has lifted this limitation for all practical purposes, at least in pulsed applications. The power which may be launched into an optical fibre is therefore limited by non-linear effects, which result from the interaction of high-intensity light with the glass forming the structure of the fibre. Optical non-linear effects occur at modest power levels in optical fibres because the guiding structure confines the optical power to a small area over very long distances, resulting in far greater interaction lengths than could be achieved with Gaussian beams in a non-guiding medium. These non-linear effects have been reported in a number of publications and are summarised below, but a more detailed review may be found (for example) in Chapter 10 of the book by K. T. V. Grattan & B. T. Meggitt (Eds.): “Optical Fiber Sensor Technology” Chapman & Hall 1995 (ISBN-0-412-59210-X).
a) Stimulated Raman Scattering (SRS):
The stimulated Raman effect results from the interaction of the incident radiation with molecular vibrations (optical phonons) and gives rise to the conversion of optical power from the incident wavelength to (in the first instance) a longer wavelength, known as the Stokes wavelength. The Stokes wavelength is separated from the incident wavelength by a frequency shift, which depends on the materials forming the fibre, but for silica-based fibres is mainly around 440 cm
−1
. Thus for incident light at 1550 nm, the first Stokes radiation appears at a wavelength of about 1663 nm.
For a probe wavelength of 1550 nm launched into a long length (of order 5 km or more) of single mode fibre of the type commonly used for telecommunications purposes, the stimulated Raman effects converts significant amounts of probe power to the first Stokes wavelength when the peak power exceeds typically 1 to 3 W, depending on the design of the fibre. If the optical power at the Stokes wavelength builds up to a sufficient level, it can itself generate light at a second Stokes wavelength and so on. Under suitable conditions, SRS can also occur at a shorter wavelength (anti-Stokes Stimulated Raman scattering), but the predominant effect is a shift to longer wavelength, which can be so efficient that most of the power of the incident light is converted to longer wavelengths.
Stimulated Raman scattering is primarily a forward-effect (i.e. the converted light travels in the same direction as the incident light) and is determined by the peak optical power. It is relatively independent of the duration of the pulse. It is also independent of the spectral width of the incident light, provided the latter falls within the broad gain spectrum of the Raman process (in the case of an incident wavelength of 1550 nm, the incident spectrum would scarcely affect the efficiency of the SRS process until it reached some 35 nm full-width at half maximum).
b) Stimulated Brillouin Scattering (SBS):
Stimulated Brillouin Scattering is caused by the interaction of the incident light with lattice vibrations (acoustic phonons), particularly those which have an acoustic wavelength similar to the incident optical wavelength. Like SRS, it results in the generation of a new wavelength, shifted with respect to the incident wavelength by a frequency equal to that of the acoustic phonons taking part in the interaction. This frequency depends on the material and the incident wavelength, but in silica-based fibres and for an incident wavelength of 1550 nm it is typically 10.7 GHz.
Unlike SRS, SBS is a very narrow-linewidth process. Thus if the incident illumination has a broader spectrum than that of the process (typically 100 MHz in silica-based fibres), then the threshold for efficient conversion is raised in proportion to the ratio of the linewidth of the source to that of the SBS process. A further difference between SRS and SBS is that the latter is primarily a backward process, i.e. the new wavelength travels in the reverse direction from that of the incident radiation. As a result, the overlap between
Font Frank G.
Mooney Michael P.
Welsh & Katz Ltd.
York Sensors Limited
LandOfFree
Optical time domain reflectometry method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Optical time domain reflectometry method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optical time domain reflectometry method and apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3042015