IEC 60793 1 42 Edition 3 0 2013 01 INTERNATIONAL STANDARD NORME INTERNATIONALE Optical fibres – Part 1 42 Measurement methods and test procedures – Chromatic dispersion Fibres optiques – Partie 1 42 M[.]
Method A, phase shift
This method describes a procedure for determining the chromatic dispersion of the categories of class B single-mode fibres B1, B2, B4, B5 and sub-categories B6_a1 and B6_a2, category
A1 graded-index multimode fibers, along with sub-categories A4f, A4g, and A4h, are analyzed over a specific wavelength range by measuring the relative phase shifts among sinusoidally modulated optical sources of varying wavelengths These sources generally include laser diodes, filtered light-emitting diodes, or filtered amplified spontaneous emission (ASE) sources The measured relative phase shifts are transformed into relative time delays, and the resulting spectral group delay data are fitted to a specific equation tailored for each fiber sub-category.
Method B, spectral group delay in the time domain
This method describes a procedure for determining the chromatic dispersion of the categories of class B single-mode fibres B1, B2, B4, B5 and sub-categories B6_a1 and B6_a2, category
A1 graded-index multimode fibres and sub-category A4f, A4g and A4h fibres with the use of a
Nd:YAG/fibre Raman laser source or multiple laser diodes operating at a number of wavelengths, both greater than and less than, the typical zero-dispersion wavelength
This method involves measuring the time delay of optical pulses traveling through a known length of fiber at various wavelengths A reference measurement is conducted using a short reference fiber, and the data from this reference is subtracted from the measurements taken from the fiber under test to determine the relative spectral group delay The resulting spectral group delay data is then fitted to a specific equation tailored for each fiber subcategory.
Method C, differential phase shift
This method describes a procedure for determining the chromatic dispersion of the categories of class B single-mode fibres B1, B2, B4, B5 and sub-categories B6_a1 and B6_a2, category
A1 graded-index multimode fibres and sub-category A4f, A4g and A4h fibres The dispersion coefficient at a particular wavelength is determined from the differential group delay between two closely spaced wavelengths
In this process, a modulated light source is introduced into the fiber being tested, allowing for a comparison of the phase of light exiting the fiber at two different wavelengths The average chromatic dispersion between these wavelengths is calculated using the differential phase shift, the wavelength interval, and the length of the fiber.
The chromatic dispersion coefficient at a wavelength midway between two test wavelengths is assumed to be the average chromatic dispersion across that interval Subsequently, the resulting chromatic dispersion data is fitted to a specific equation tailored for each fiber subcategory.
Category A1 and sub-category A4f, A4g and A4h multimode fibres
For category A1 and sub-categories A4f, A4g, and A4h multimode fibers, the reference test method (RTM) for resolving disputes is method B, which measures spectral group delay in the time domain.
Class B single-mode fibres
For class B single-mode fibers, including categories B1, B2, B4, B5, and sub-categories B6_a1 and B6_a2, the recommended testing method is method A, known as phase shift If method A is unavailable, method C, which involves differential phase shift, can be utilized to address any disputes.
General
The following apparatus is common to all measurement methods Annexes A, B and C include layout drawings and other equipment requirements that individually apply for each of the methods, A, B and C respectively.
Launch optics
To ensure accurate measurements, the output from signal sources must be consistently coupled to either the fibre under test or the reference fibre, maintaining a constant physical path length for each source This consistency prevents any changes in the relative phases of the sources caused by variations in path length Appropriate devices for this purpose include multichannel single-mode optical switches and demountable optical connectors.
For measurement of category A1, and sub-category A4f, A4g, A4h multimode fibre, launch conditions shall comply with method A, Impulse response, of IEC 60793-1-41.
High-order mode filter (single-mode)
To measure single-mode fiber effectively, it is essential to employ a technique that eliminates high-order propagating modes within the desired wavelength range A practical example of a high-order mode filter is a single loop with a sufficiently small radius, which can shift the cut-off wavelength below the minimum wavelength of interest.
Input positioning apparatus
To ensure effective coupling of the specimen's input to the light source, various methods can be employed, including x-y-z micropositioner stages and mechanical coupling techniques like connectors, vacuum splices, and three-rod splices It is crucial that the fiber's position remains stable throughout the testing period.
Output positioning apparatus
To ensure effective coupling of guided optical power to the system detector, it is essential to position the output end of the specimen appropriately This coupling can be achieved through the use of lenses or by establishing a mechanical connection to a detector pigtail.
Computation equipment
A digital computer may be used for purposes of equipment control, data acquisition, and numerical evaluation of the data
Specimen length
Methods A, B, and C require the specimen to be a fibre or cable of known length sufficiently long to produce adequate phase measurement accuracy A typical minimum length is 1 km
Because sub-category A4f, A4g and A4h fibres have higher loss than category A1 fibres, for these A4 fibres a minimum length of 100 m is acceptable
NOTE Reproducibility is affected when using shorter measuring length Longer lengths generally yield better reproducibility.
Specimen end face
Prepare a flat end face, orthogonal to the fibre axis, at the input and output ends of each specimen.
General
The calculation of relative delay appropriate for each method is given in Annexes A, B and C, respectively
This clause outlines the numerical fitting applicable to all methods for the normalized spectral group delay data, denoted as τ(λ), which is dependent on the wavelength (λ) measured in nanometers (nm) For further details, refer to Annex D The normalized spectral group delay data fit is expressed in picoseconds per kilometer (ps/km).
D(λ ) is the chromatic dispersion coefficient, with D(λ ) = dτ ( λ ) /dλ (ps/(nm × km) ) λ 0 is the zero-dispersion wavelength (nm) τ(λ 0 ) is the relative delay minimum at the zero-dispersion wavelength (ps/km)
S(λ) is the dispersion slope, with S(λ ) = d D(λ ) /dλ (ps/(nm 2 × km))
S 0 is the dispersion slope at the zero-dispersion wavelength (ps/(nm 2 × km))
NOTE 1 τ ( λ ) and D( λ ) may either be direct measurements or the result of fitting the direct measurements to a specified function
In data fitting, the parameters on the right side of the equation are optimized to minimize the sum of squared errors relative to direct measurements Once these parameters are established, the resulting expression is utilized to calculate various other parameters.
NOTE 3 The fit parameters are given as the variables A, B, C, D, or E, see also Annex D.
Category A1 and sub-category A4f, A4g, A4h multimode fibres and category B1.1, B1.3 and sub-category B6_a1 and B6_a2 single-mode fibres
B1.3 and sub-category B6_a1 and B6_a2 single-mode fibres
The following applies to category A1 and sub-category A4f, A4g and A4h multimode fibres, and to category B1.1, B1.3 and sub-category B6_a1 and B6_a2 single-mode fibres around
The delay or dispersion data fit shall be fitted with the 3-term Sellmeier fit type, see Annex D
Calculations for the chromatic dispersion coefficient D(λ), the zero-dispersion wavelength λ 0 and the dispersion slope at the zero-dispersion wavelength S 0 are shown in Annex D
In the 1 550 nm region only, the chromatic dispersion can be approximated as a linear function with wavelength (quadratic fit type to the delay data), see Annex D.
Category B1.2 single-mode fibres
The following applies to category B1.2 single-mode fibres
For wavelength intervals up to 35 nm in the 1,550 nm region, a quadratic fit is permissible, but it should not be used to predict chromatic dispersion outside this range For longer wavelengths, the 5-term Sellmeier fit or a 4th order polynomial fit is recommended, with the caveat that these methods are not suitable for the 1,310 nm region.
Calculations for the chromatic dispersion coefficient D(λ) and the dispersion slope S(λ) are shown in Annex D.
Category B2 single-mode fibres
The following applies to category B2 single-mode fibres
For wavelength intervals up to 35 nm, a quadratic fit is permissible in the 1,550 nm region, depending on accuracy needs However, the fitted equation should not be applied to predict chromatic dispersion at wavelengths beyond the fitting range.
For longer wavelength intervals, it is advisable to use either the 5-term Sellmeier fit or the 4th order polynomial fit However, these methods are not suitable for the 1,310 nm region.
The corresponding chromatic dispersion coefficient D(λ), the zero-dispersion wavelength λ 0 and the dispersion slope at the zero-dispersion wavelength S 0 are shown in Annex D.
Category B4 and B5 single-mode fibres
The following applies to category B4 and B5 single-mode fibres
For extended wavelength intervals greater than 35 nm, it is advisable to utilize either the 5-term Sellmeier fit or the 4th order polynomial fit It is important to note that the fitted equation should not be applied to predict chromatic dispersion for wavelengths outside the fitting range.
For B4 fibres only, the quadratic fit type may be used in case of a short wavelength interval
(≤ 35 nm) The fit type should not be used to predict chromatic dispersion at wavelengths outside the range used for the fit
The corresponding chromatic dispersion coefficient D(λ) and the dispersion slope S(λ) are shown in Annex D
Report the following information with each measurement
– date and title of measurement;
– equation(s) used to calculate the results;
– length of specimen used for length normalization;
– measurement results as required by the detail specification
The detail specification may require essential information such as the dispersion coefficient values measured at specific wavelengths, the minimum and/or maximum dispersion over a defined range of wavelengths, and the zero-dispersion wavelength along with the dispersion slope at that wavelength.
The following information shall be available upon request
– description of optical source(s) and measurement wavelengths used;
– description of signal detector, signal detection electronics and delay device;
– description of computational techniques used;
– date of latest calibration of measurement equipment
The detail specification shall specify the following information:
– (sub) category of fibre to be measured;
– any deviations to the procedure that apply
Requirements specific to method A, phase-shift
A stable light source is essential for accurate measurements, maintaining consistent position, intensity, and wavelength over a sufficient duration Options for light sources include multiple laser diodes, wavelength-tunable laser diodes, light-emitting diodes, or broadband sources such as a Nd:YAG laser with a Raman fiber or an ASE source, depending on the required wavelength range for the measurement.
The wavelength introduced into the tested fiber can be chosen through various methods, including optical switches, monochromators, dispersive devices, optical filters, optical couplers, or by adjusting the laser, depending on the light sources and measurement configuration This wavelength selection can occur either at the fiber's input or output.
In a three-wavelength system used for measuring category B1 fibres, where the source wavelengths encompass the zero-dispersion wavelength, λ₀, any tolerance or instability, δλ, in the center wavelength can result in maximum measurement errors of 3δλ for λ₀.
Maximum errors in dispersion slope, S 0 , are directly proportional to δλ/∆λ (where ∆λ = source wavelength spacing) and will be approximately 0,012 ps/(nm 2 × km) for δλ/∆λ = 1 nm/30 nm
To achieve errors smaller than the maximum allowable limits, it is essential to select optical sources with an average wavelength that closely matches the expected wavelength (\( \lambda_0 \)) of the specimen Additionally, utilizing more than three wavelengths can further enhance accuracy.
For effective use of laser sources, a temperature-controlled, single longitudinal-mode laser diode with output power stabilization, such as PIN photodiode feedback, is generally adequate However, an additional laser may be necessary for establishing a reference link in field measurement setups.
The spectral width of the source, as measured in the specimen, shall be less than or equal to
10 nm at 50 % power points (FWHM)
Multiple laser diodes Optical switch
Test sample or calibration fibre
Fibre reference link (equipment not positioned adjacent to each other)
Electrical reference (equipment positioned adjacent to each other)
Fibre reference link (equipment positioned adjacent to each other)
Figure A.1 – Chromatic dispersion measurement set, multiple laser system (typical)
Delay - ps/km Data fit - ps/km Dispersion - ps/(nm × km)
R el at iv e d el ay - ps/ km D is per si on - ps /(nm × km)
Figure A.2 – Typical delay and dispersion curves
The modulator will amplitude modulate light sources to generate a waveform with a primary Fourier component, allowing for sinusoidal, trapezoidal, or square wave modulation Additionally, it is essential that the frequency stability achieves a minimum of one part in 10^6.
To avoid ambiguities in measuring phase shifts of 360(n) degrees, where n is an integer, it is crucial to track 360° phase changes Alternatively, selecting a modulator frequency that is sufficiently low can help ensure that the relative phase shifts remain below this threshold.
360° Determine the maximum frequency for a 360° shift for category B1 fibres as: f max 2 1 j
S 0 (A.1) where f max is the maximum frequency for a 360° shift for category B1 fibres (MHz);
L is the maximum expected specimen length (km);
The expected typical dispersion slope at the zero-dispersion wavelength, denoted as \$S_0\$, is measured in ps/nm² × km The zero-dispersion wavelength is represented by \$\lambda_0\$ in nanometers In the measurement process, the wavelength pair \$\lambda_i\$ and \$\lambda_j\$ is utilized to minimize \$f_{max}\$.
The frequency of the modulator shall be sufficiently high to ensure adequate measurement precision
The precision of test systems is influenced by their parameters, as demonstrated in a three-wavelength system for category B1 fibers In this setup, the maximum errors are 0.0012 ps/nm² × km for S₀ and 0.4 nm for λ₀, given that the minimum modulator frequency, fₘᵢₙ, is set at 72 MHz.
L (A.2) where f min is the minimum modulator frequency (MHz);
∆φ is the overall measurement equipment phase instability (degrees);
L is the minimum expected specimen length (km);
∆λ is the average wavelength spacing between adjacent sources (nm)
Hence for ∆φ = 0,1°, L = 10 km, and ∆λ= 32 nm, a minimum frequency of approximately
NOTE 1 Equation (A.2) above was developed by repeatedly solving for λ 0 and S 0 in the time-delay Equation (2) of Clause 6 with various values of wavelength spacing and phase instability
To achieve errors smaller than the maximum specified, it is essential to select sources with an average wavelength that closely matches the expected wavelength (\(\lambda_0\)) of the specimen Additionally, utilizing more than three wavelengths can further enhance accuracy.
The phase modulation at each light source may be adjustable to facilitate measurement-set calibration
A.1.4 Signal detector and signal detection electronics
To measure specific wavelengths effectively, utilize an optical detector that is sensitive across the desired range, paired with a phase meter Enhancing the detection system's sensitivity can be achieved by incorporating an amplifier A standard setup often consists of a PIN photodiode, a field-effect transistor (FET) amplifier, and a vector voltmeter.
The detector-amplifier-phase meter system is designed to respond exclusively to the fundamental Fourier component of the modulating signal, ensuring a consistent signal phase shift across varying optical power levels The range of received optical power can be adjusted using a variable optical attenuator.
To accurately measure the phases of signal sources, it is essential to provide a reference signal that shares the same dominant Fourier component as the modulating signal This reference signal must be phase-locked to the modulating signal and is usually derived from it.
Reference signal configurations can vary based on the setup and testing conditions In laboratory tests or calibration scenarios, signal sources and detectors positioned adjacent to each other can utilize an electrical connection between the signal generator and the reference port of the phase meter Alternatively, an optical splitter can be employed before the specimen for equipment that is also positioned closely For field testing of optical cables, where the sources and detectors are not adjacent, an optical link is typically used, consisting of a modulated light source, fiber, and detector similar to those used for the specimen Additionally, wavelength division multiplexing can transmit a reference signal on the fiber under test during field testing.
Test sample or calibration fibre
Electrical reference (equipment positioned adjacent to each other)
Fibre reference link (equipment not positioned adjacent to each other)
Figure A.3 – Chromatic dispersion measurement set, LED system (typical)
Insert the reference fibre (6.3) into the measurement apparatus, and establish a reference signal (A.1.5) Measure and record the phase, φ in (λ i ), for each signal source
If the signal sources can be adjusted in phase, the phases of all sources can be equalized with the reference fiber in place Subsequently, specimen measurements should be conducted as outlined in section A.2.2 In this scenario, the value of φ in (λ i) is set to 0 for the calculations in A.3.1.
Insert the specimen into the measurement apparatus, and establish a reference signal (see
A.1.5) Measure and record the phase, φ out (λ i ), of each signal source