IEC 60793-2-10, Optical fibres – Part 2-10: Product specifications – Sectional specification for category A1 multimode fibres IEC 60793-2-20, Optical fibres – Part 2-20: Product specif
General
This article outlines two methods (A and B) for generating images of fibers in a plane perpendicular to their propagation axis The resulting images are analyzed to express the fiber's geometry, as detailed in Annexes C, D, and E Both methods can yield one-dimensional or two-dimensional images, with the latter providing richer information and more comprehensive geometric details Notably, one-dimensional scans cannot accurately assess non-circularity or concentricity errors, highlighting the advantages of two-dimensional imaging for precise analysis.
The image analysis involves two key steps: first, identifying the delineation of the body of interest within the image, and second, condensing these delineation points into geometric parameters such as diameter, non-circularity, and center If both the cladding and core are measured, concentricity error can also be assessed Methods applicable to both cladding and core for all fiber types are detailed in Annex D, while Annex E outlines a specific method for the core body of class A fibers.
This standard addresses a range of needs, and as such, allows for a range of for data collection and reduction The specific limitations and uses of these approaches are discussed below.
Scanning methods
General
Sampling a two-dimensional body using only one dimension presents significant limitations While ideal fibers are perfectly circular with concentric core and cladding, real fibers often exhibit non-circularity and concentricity errors These characteristics cannot be accurately assessed through one-dimensional scans, which may lead to under- or over-estimation of the average diameter of a noncircular body However, one-dimensional scanning can be effective for fibers with minimal non-circularity and concentricity errors, and it is frequently employed to measure the core diameter of class A fibers.
One-dimensional scan sources of error
Figure 1 demonstrates the error that arises when the sampling axis is not aligned with the center of the body, leading to an underestimation of the body's diameter This misalignment results in a second-order error.
If a body is non-circular, a one-dimensional scan will not fully describe the body’s shape
Sampling a body in one dimension often leads to inaccuracies in estimating its average diameter While it might seem that sampling in two orthogonal axes (X and Y) could resolve this issue, this approach is typically inadequate.
Figre 2a – Major diameter Figure 2b – Average diameter
Figure 2 – Scan of a non-circular body
Errors in sampling an elliptical body on one or two axes are illustrated in Figure 2 When the ellipse's major diameter aligns with the X axis, sampling solely in X leads to an overestimation of the average diameter, while sampling only in Y results in an underestimation Sampling on both axes accurately characterizes the body, revealing both its average diameter and non-circularity However, in the case of average diameter, sampling on either axis yields a similar, approximately correct diameter, potentially misleading one to believe the body is circular Analyzing ±45° scans provides accurate measurements of non-circularity and diameter, but the optimal angular scan angles are unknown in advance At orientations other than –45° and +45°, the average diameter is measured correctly, but the circularity is underestimated.
If a single axis is scanned, the core’s centre relative to the cladding centre cannot be known
Scanning along two orthogonal axes offers a reliable estimate of the core's center However, this estimate diminishes in accuracy if the core is scanned along a chord that is distant from its center Additionally, if the core is much smaller than the cladding and is notably non-concentric, there is a risk that one or more scans may completely overlook the core.
Multidimensional scanning
As suggested in 5.2.2.2 and 5.2.2.3, the estimation of the geometry of the fibre can be improved by scanning on two orthogonal axes Combining scans over more than two angles
Acquiring data at various angles, such as 0°, 45°, 90°, and 135°, enhances the accuracy of estimates This can be achieved by rotating the fiber in its holding chuck or through the scanner's mechanics if designed for such functionality It is crucial that all angular scans maintain a single frame of reference, as a common origin is necessary to avoid introducing errors.
If the scanner is capable of motion on two orthogonal axes, then it is possible that a two- dimensional image of the fibre may be constructed by performing a raster scan
Measurement of the transmitted near-field using grey-scale video is inherently a raster scan
Data reduction
Simple combination of few-angle scan sets
To effectively reduce data sets with limited angular orientations, simple data reduction techniques are often adequate The diameter of each body can be calculated by averaging the diameters from various angular scans, while non-circularity is assessed using the maximum and minimum diameters from the measured angles Additionally, when both cladding and core measurements are available, the concentricity error can be identified from the angle that exhibits the most significant centration error For further details, refer to Annex D.
Ellipse fitting of several-angle or raster data sets
When numerous data points are obtained from a scan set, particularly with multiple angles or raster scanning, edge tables can be modeled using elliptical shapes The methodology for fitting a body's edge table is detailed in Annex E, following the procedures outlined in Annex D.
For both the cladding and the core for all fibre categories, ellipse fitting is the reference method
The reference test method (RTM) utilizes the video grey-scale transmitted near-field technique for all fibre categories Data analysis involves boundary detection and ellipse fitting to streamline edge tables into geometric representations For insights on reference sample lengths across all fibre classes, consult Annexes A and B, while Annex C provides details on the decision threshold factor k specifically for class A fibres.
Annexes A and B include layout drawings and other equipment requirements for each of the
Specimen length
Annexes A and B specify the required sample lengths for their respective methods.
Specimen end face
Ensure that both the input and output ends of each specimen have a clean, flat end face that is perpendicular to the fibre axis, as any deviation from this perpendicularity can compromise the accuracy of measurements.
End angles less than 1 ° are recommended
See Clause B.2 for the tighter requirements on end faces when using Method B
Use the procedures given in IEC 61745 for calibration Annexes A and B document the procedures for Methods A and B, respectively
Refer to Annexes C, D and E for details regarding the calculations
The following information shall be provided with each measurement:
– date and title of measurement;
– identification and description of specimen;
– measurement results for each parameter specified (see the applicable annex)
The following information shall be available upon request:
– measurement method used: Method A or B;
– arrangement of measurement set-up;
– details of measurement apparatus (see applicable annex);
– relative humidity and ambient temperature at the time of the measurement;
The detail specification shall specify the following information:
– type of fibre to be measured;
– any deviations to the procedure that apply
Requirements specific to Method A – Refracted near-field
Introductory remarks
The refracted near-field measurement technique accurately assesses the variation in refractive index within the fiber's core and cladding This method can be calibrated to provide absolute refractive index values and is effective for profiling both single-mode and multimode fibers.
A refracted near-field measurement assesses the radial dependence of relative index variations in a fiber by scanning a light spot across its end-face By injecting a theoretical ray of light at an angle exceeding the fiber's maximum numerical aperture and measuring its exit angle, changes in index can be detected However, due to the impossibility of generating an ideal ray and the fiber's dimensions being on the order of 100 optical wavelengths, an integral approach using an angular bundle of rays is employed A small light spot with a numerical aperture greater than the fiber's is scanned at a normal angle of incidence, and the light cone exiting the fiber is sampled at high angles The total power in this sampled region is determined based on the radial location of the launch spot As light interacts with local index differences in the fiber, it refracts, altering its exit angle Light passing through the core and cladding exits at shallower angles than light passing solely through the cladding, resulting in lower detected power from the core region compared to the cladding Consequently, the relative power at a specific scan position is directly proportional to the fiber's index at that location.
Apparatus
Typical arrangement
See Figures A.1 and A.2 for schematic diagrams of the test apparatus.
Source
Provide a stable laser giving a few milliwatts of power in the TEM 00 mode
A HeNe laser with a wavelength of 633 nm is adequate for geometrical measurements However, if the index needs to be measured, a correction factor may be necessary to adjust the results for different wavelengths.
A quarter-wave plate is used to convert linear polarization into circular polarization, enabling the generation of a time-averaged signal that remains unaffected by polarization variations This is particularly important because the reflectivity of light at an air-glass interface is significantly influenced by both the angle of incidence and the polarization state.
If necessary, place a spatial filter, such as a pin-hole, at the focus of the microscope objective.
Launch optics
To achieve optimal performance, arrange the launch optics using a high magnification, high numerical aperture microscope objective to overfill the numerical aperture (NA) of the fiber This setup focuses a beam of light onto the flat end of the fiber, ensuring that the optical axis of the beam is aligned within 1° of the fiber's axis Additionally, to enhance spatial resolution, the size of the focused spot should be minimized, ideally to less than 1.5 µm.
XYZ positioner (scanning stage)
The launch optics or cell must be installed on a three-axis positioner that allows for movement exceeding the anticipated fiber diameter The focus axis (Z) resolution should be high enough to maintain a sharp focus on the fiber end-face, ensuring that the instrument's spatial resolution is not significantly affected Additionally, the resolution of the other two axes is also crucial for optimal performance.
(X and Y) shall be smaller than half the focused spot size
Figure A.1 – Refracted near-field method – Cell
Blocking disc
The blocking disc is designed to allow only light that refracts out of the fibre without internal reflection to reach the detector While extending the fibre can help by bending it out of the optical path and guiding some light away, this alone is inadequate Partial internal reflection at the cladding/oil interface can cause some light to reflect back into the fibre Consequently, when non-refracted light reaches the detector, it results in an increase in measured power, leading to a negative error in index determination.
Guided modes And leaky modes Disc
The blocking disc is designed to obstruct a specific angular cone of light from reaching the detector, ensuring that most non-refracted light is blocked while preserving sufficient refracted light to maintain optimal signal-to-noise performance Generally, the numerical aperture (NA) of the subtended cone is chosen to be about the light source's NA at the fiber end-face, divided by the square root of 2.
Collection optics and detector
Measuring the total power of light passing through a blocking disc is crucial This can be achieved using large condenser lens systems, parabolic and elliptical mirrors, and integrating spheres, among other methods A practical setup must balance the size of the detector with optical complexity, ensuring accurate measurement of total light power up to the numerical aperture (NA) launched into the fiber Additionally, the detector's noise and dynamic response should not significantly affect the measurement accuracy.
The detector must be sensitive to the light source's wavelength and exhibit linearity across the anticipated optical power levels To enhance the detector's signal and automatically measure relative differences during stage scanning, amplifiers and data converters are usually integrated with the detector.
Computer system
A computer collects data by managing the positioner and digitizing the detector signal After data collection, it converts the detector signal into index difference or absolute index through proper calibration.
Immersion cell
The immersion cell surrounds the optical fibre, ensuring that light exiting the fibre encounters a sufficiently high refractive index to prevent any light from reflecting back into the fibre It is crucial for the optical media around the cladding to have a higher refractive index than the cladding itself, which is achieved using index matching oils The design of the immersion cell can vary, provided it does not significantly impact the refraction of light into the collection optics.
Sampling and specimens
The length of a fiber sample is determined by the design of the instrument It is crucial to ensure that the output end of the fiber, which is not positioned in the scanning plane, does not couple light into the detector.
Remove all fibre coatings from the section of fibre to be immersed in the liquid cell.
Procedure
Load and centre the fibre
To determine the rough fibre center, \(X_f\) and \(Y_f\), place the fibre sample in the cell and utilize methods such as back illumination with a tungsten lamp or scanning techniques.
XY stage to search for the fibre Adjust the stage to centre and focus the source spot on the fibre end
Center the disc on the output cone as specified by the instrument design For class A multimode fiber, ensure the disc is positioned on the optical axis to effectively block the leaky modes.
B and C single-mode fibres, position the disc to give optimum resolution
Once the fibre is centred and the disc is aligned, either line scans or a complete raster scan can be performed.
Line scan
To conduct a scan of the stage at a specific angle of interest, φ, it can be performed at 0° using only the X stage, at 90° using only the Y stage, or at any suitable angle utilizing both stages, with the stage resolution and desired scan resolution dictating the feasible angles The scan range must extend beyond the cladding on both sides of X f and Y f Additionally, the radial spacing of the scan should be chosen to adequately sample the index variation, ensuring accurate determination of the fiber’s geometry A total of nS power readings are collected during this process.
P i is the set of detected power readings; x i is the set of radii where the power readings were collected.
Raster scan
To accurately determine the fiber's geometry, scan the stage in a raster pattern across both the X and Y axes, ensuring the range encompasses the cladding The spacing for the X and Y scans must be chosen to adequately sample the index variation A series of power readings will be collected during this process.
P j,I is the set of detected power readings, x i is the set X-axis points where the power readings were collected, y j is the set Y-axis points where the power readings were collected.
Calibration
The angle of the light cone changes based on the refractive index at the fiber's entry point, affecting the power passing through the disc By removing the fiber and knowing the liquid index and cell thickness, this angle change can be simulated by adjusting the disc along the optical axis Moving the disc to specific positions allows for scaling the profile in terms of relative index, which helps determine the instrument's delta calibration factor, K ∆ Absolute indices, n 1 and n 2, can only be accurately determined if the cladding index or liquid index at the measurement wavelength and temperature is known.
The geometric scaling factors, S X and S Y (measured in micrometres per stage step), must be established for the scanning stage This can be achieved by scanning a traceable artefact like a chrome-on-glass reticule, certifying the stage micrometers or indexers, or utilizing other suitable methods.
A multi-index calibration artefact, which may be made available from national standards institutes, may also be used to determine K ∆ , S X and S Y
Index of refraction calculation
Determine relative index profile, ∆ i (or alternatively ∆ i,j for a raster scan)
The reference power level, denoted as P ref, establishes the point in the profile where the index difference is zero This reference can be chosen at any convenient location within the profile or set as an instrument parameter, and its value does not influence the subsequent calculations.
Figure A.3 – Typical index profile line scan of a category A1 fibre
Figure A.4 – Typical raster index profile on a category A1 fibre
Figure A.3 and A.4 show typical index profile data of a category A1 fibre Figure A.4 expresses the index of refraction as a grey-level of intensity, with whiter colours indicating higher index
Index of ref rac tio n
Calculations
Refer to Annexes C, D and E to reduce the index scan set to geometry, substituting ∆ for I.
Results
The following parameters may be determined from the measurement:
– core diameter (class A multimode fibres only);
– core non-circularity (of type A fibre);
In addition to the results listed in Clause 11, and depending on the specification requirements, the following information shall be provided on request:
– profiles at specific angles calibrated for a given wavelength;
– equipment arrangement and wavelength correction procedure
Requirements specific to Method B – Transmitted near-field
Introductory remarks
The transmitted near-field method is used to ascertain the geometric parameters of class A multimode fibers and class B and C single-mode fibers by examining the optical power density in relation to position on a cross-section at the fiber's end This annex outlines two techniques that focus on analyzing the near-field image of an optical fiber's end-face.
– the video grey-scale technique, employing a video camera to analyse the image two- dimensionally;
– the mechanical scan technique, in which one or more one-dimensional scans of the image are acquired for analysis
The video grey-scale technique is the reference test method (RTM)
One-dimensional mechanical scanning is commonly employed to measure the core diameter of class A multimode fibers However, as noted in Clause 5, relying solely on one-dimensional scans has its limitations To address these challenges, multiple one-dimensional scans can be integrated using the data reduction techniques outlined in Annexes C and D, although this approach increases measurement time and complexity Generally, one-dimensional near-field scanning is utilized for accurately determining the core diameter of class A multimode fibers.
Apparatus
Typical arrangement
Figures B.1 and B.2 are examples of apparatus configuration for the two techniques
Figure B.1 – Typical arrangement, grey scale technique
Figure B.2 – Typical arrangement, mechanical scanning technique
Light sources
Use suitable incoherent light sources for the illumination of the core and the cladding, adjustable in intensity and stable in intensity over a time period sufficient to perform the measurement
Class A multimode fibres require core geometry to be determined using incoherent illumination that fully fills the core at the operational wavelength In contrast, for Class B and C single-mode fibres, while the core center is identified using this method, core diameter and circularity are not specified Consequently, the illumination requirements for Class B and C fibres are less stringent, allowing for any convenient wavelength that adequately overfills the few modes propagating in the fibre.
C core centre does not substantially change with wavelength even when more than one mode group propagates in the core
The geometry of category A1, A2, and A3 multimode fibres is defined using a core illumination center wavelength of 850 nm ±10 nm, unless specified otherwise For category A4 fibres, the geometry is determined at a core illumination center wavelength of 650 nm ±10 nm, unless stated otherwise Additionally, the full-width-half-maximum width of the core illuminators for all class A fibres must exceed a specified threshold.
10 nm and less than 50 nm
Currently, all class A fibre specifications are under revision to incorporate the center wavelength for core geometry determination Once the updated specifications are released, the previous information will be disregarded, and the new product specifications will take precedence.
Cladding can be illuminated in two ways: either by reflecting light off the cleaved end-face of the fiber, which keeps the surrounding air dark, or by flooding the surrounding air with light while leaving the cladding unlit.
The IEC wavelength is not critical; however, its relationship to the core illuminator's wavelength is essential due to the dispersion of the magnifying optics Choosing a wavelength that is similar or falls within the performance window of the optics will help maintain focus on the core while the cladding is in focus.
When measuring the core diameter of class A fibers, the cladding is typically not illuminated The video grey-scale technique allows for the determination of core diameter and non-circularity using an image or scan without cladding illumination, while a separate illuminated image of the cladding can be utilized to assess additional parameters.
Fibre support and positioning apparatus
Ensure a stable support system for the specimen's input and output ends, such as a vacuum chuck Position these supports on devices that allow precise alignment of the fiber ends within the input and output paths Utilizing three-axis translation stages for the support apparatus can enhance convenience, as these stages may also function as the scanner in certain mechanical scanning techniques.
Cladding mode stripper
Use devices that effectively strip cladding mode light from the specimen close to the fiber input and output ends If the fiber being tested has a coating layer in contact with the cladding, and this layer has a higher refractive index than the glass, it will function as a cladding mode stripper.
Detection
For effective measurement precision, the detection system must maintain sufficient linearity Typically, PIN photodiodes in photovoltaic mode and modern camera sensors meet this criterion; however, careful selection and usage are essential Performance can be compromised by high illumination levels, as well as by inadequately designed conditioning electronics and digitization systems.
The video grey-scale technique utilizes a video camera to capture a magnified near-field image, with a video digitizer often integrated for image digitization and analysis The resulting digitized output is represented as a pixel array of near-field intensities, denoted as I(r,c), organized in N Row rows and N Col columns Both CCD and CMOS imaging sensors are suitable for this application, and it is essential that the effective pixel size meets specific requirements.
Systematic errors in the detection system can compromise measurement precision, with examples including the geometric uniformity of digitized images and the linearity of the detector/digitizer in response to variations in optical intensity It is crucial to address these potential errors, and IEC 61745 outlines the necessary methodology for assessing their magnitude.
B.2.5.3 Mechanical scan detector and scanner
The mechanical scan detector utilizes a fixed-aperture design and a scanning system to capture image intensity based on position It enables the scanning of the focused image of the fiber's near-field pattern, with a calibration that ensures the relative radial position is known By employing a high-resolution mechanical scanner, either the fiber or the imaging system and detector can be moved together Alternatively, scanning the detector in the image plane allows for the use of a lower-resolution mechanical scanner Regardless of the method, the mechanical scanner must maintain sufficient linearity to meet the required measurement precision.
To ensure compliance with Equation (B.1), the effective aperture of the detector must be limited A detector with a small active area, such as a 20 µm diameter detector paired with a 40X imaging system, can meet this requirement The aperture can be restricted by using an optical fiber with a sufficiently small core diameter, ensuring that its input end is in focus at the image plane and its output is connected to the detector Alternatively, a mechanical pinhole can be utilized, with relay optics employed to project the backside of the pinhole onto the optical detector.
Magnifying optics
B.2.6.1 Optical imaging system general information
To effectively analyze the specimen, it is essential to implement an imaging system that enhances the near-field output image, allowing for precise scanning This system must possess a numerical aperture that exceeds that of the fiber core being measured.
The numerical aperture of the imaging system affects the resolving power of measurement, and thus shall be compatible with the measuring accuracy, and not lower than 0,3
In video grey-scale techniques, the largest dimension of the pixels or the size of the detector (or pin-hole) in mechanical scan techniques must be significantly smaller than the magnified near-field image, ideally less than half of the system's diffraction limits.
1,22Mλ d≤ (B.1) where d is the pixel size of the camera, or the detector (pin-hole) size in àm;
M is the approximate magnification of the optical system; λ is the (lowest) test wavelength in àm;
NA is either the numerical aperture of the fibre’s core for core diameter-only measurements of class A fibres, or, for all other applications, the numerical aperture of the objective
(assuming the cladding illuminator completely fills the optical system in NA)
To ensure accurate imaging, it is essential to calibrate the optical system alongside the scanning system to determine the overall system magnification The magnification indicated on the microscope objective is not sufficient, as the scanning system's pixel spacing or mechanical step size also contributes to the total magnification and must be calibrated accordingly.
B.2.6.2 Considerations for the video grey-scale technique
When employing the video grey-scale technique, it is crucial to adjust the magnification to adequately fill the sensor area of the video camera with the image of the object being measured This means focusing on the cladding of the fiber when both cladding and core measurements are needed, or solely on the core for core-only measurements Additionally, it is important to ensure that the effective pixel size meets the criteria outlined in Equation (B.1).
Both the X and Y axes shall be calibrated, and these calibrations are generally independent
IEC 61745 provides the methodology required to perform this calibration The resultant calibration factors, in units of micrometres per pixel are S X and S Y
B.2.6.3 Considerations for the mechanical scanning technique
When employing the mechanical scanning technique, it is essential to choose the appropriate magnification of the imaging system and the size of the detector aperture to comply with Equation (B.1) Additionally, the scanner resolution, defined as the minimum step size, must not exceed half the diameter of the detector aperture.
The scanner must undergo calibration to determine the calibration factor, denoted as S X, in micrometres per step This can be achieved using a calibration artifact that is traceable to a national laboratory, such as a chrome-on-glass ruler or a grid of dots If both axes of the scanner are utilized, each axis must be calibrated separately, resulting in two independent calibration factors, S X and S Y.
Video image monitor (video grey-scale technique)
Utilize a video image monitor to showcase the detected image, which usually features a pattern like cross-hairs to help the operator center the specimen's image Computer-controlled alignment and focusing may be employed, and frequently, the monitor is integrated with the computer's display for enhanced functionality.
Computer
Use a computer to acquire the data, perform the analysis and produce the appropriate reports.
Sampling and specimens
To ensure accurate cladding measurements, prepare the specimen with clean, smooth fibre ends that are perpendicular to the fibre axis, maintaining an end angle of less than 1° from normal Minimize end damage to enhance measurement accuracy and precision, and avoid sharp bends during fibre deployment.
For class A multimode fibres, the standard sample length is 2 m ±0.2 m, except for the bend-insensitive A1a fibres (A1a.1a, A1a.2a, A1a.3a), which use a reference test length of 100 ±2 m for dispute resolution Shorter lengths may be used for routine measurements If a different reference length is specified, 2 m measurements can be mapped to the reference length, as detailed in Annex F.
Currently, all class A fibre specifications are undergoing revisions to incorporate the reference length for core geometry determination Once these updated specifications are released, the previous information will be disregarded, and the details in the product specifications will take precedence.
There is no length restriction for class B and C single-mode fibres Typically a 2 m sample length can be used.
Procedure
Equipment calibration
Artefacts traceable to a national standards laboratory shall be used to calibrate the apparatus.
Measurement
B.4.2.1 Measurement by the video grey-scale technique
To ensure the specified launch conditions, align the specimen at the input end and focus the near-field image of the output end on the camera, utilizing either automated or manual methods Optimize the signal-to-noise ratio by adjusting the core and cladding illuminators, while preventing pixel saturation.
Digitized video data is recorded as an array of pixel intensities, denoted as I The spacing parameters for the X and Y axes, represented by δX and δY, correspond to the magnification calibration parameters, SX and SY, respectively.
B.4.2.2 Measurement by the mechanical scan technique
Prepare and secure the specimen as instructed, ensuring the output end is adjusted for scanning the magnified image Focus the image on the scanning aperture's plane and center it to align the core's center with the expected position Optimize the signal-to-noise ratio by adjusting the illuminator(s) Typically, mechanical scanning is used solely for determining category A1 fibre core geometry, utilizing only a core illuminator without illuminating the cladding.
Scan the near-field image, and record the intensities, I, and their associated positions, x
Figure B.3 – Typical 1-D near-field scan, category A1 core
B.4.2.2.2 Combinations of one-dimension scans at a set of angles
Acquire scans at various angles, φ, ensuring they share a common origin For multimode core scans or those including cladding, each scan must pass through the center of the core or cladding, which may require realigning the scanner for each orientation.
Acquire scans as described in B.4.2.2.1 at a set of lines perpendicular to the axis scanned in
B.4.2.2.1 at raster positions recorded in y The covered raster distance should be the same as the covered scan distance
Figure B.4 − Typical raster near-field data, category A1 fibre
Calculations
Refer to Annexes C, D and E to reduce the near-field intensity data to geometry.
Results
In addition to the results listed in Clause 11, and depending on the specification requirements, the following information shall be provided on request:
– detector type and aperture size (single near-field scan technique only)
Edge detection and edge table construction
Introductory remarks
Edge detection is a crucial process in transforming RNF or TNF data to define the geometry of a body By analyzing these boundaries, we can further refine the geometry through various transformations, including simple differences.
The article discusses the measurement of diameter and the averaging of the center between two diametrically opposed edges, as well as the fitting of ellipses to sets of edges It highlights that Class A, B, and C optical fibers consist of two main components: the core and the cladding The edge detection techniques mentioned assume that these components are roughly circular and nearly concentric.
The core boundary shall be determined by the decision-level technique (described in
Clause C.2 outlines the core boundary decision-level value for all class A multimode fibres, while class B and C single-mode fibres do not have a specified value, although recommended values are provided The cladding boundary can be determined using the decision-level technique, along with other methods that utilize various spatial filters in one or two dimensions, though these methods are not detailed in this standard It is important to note that for the video gray-scale RNF technique mentioned in Annex B, the edge detection method used to identify the cladding boundary must be consistent with the technique employed to calibrate the cladding diameter against a known diameter artifact.
Boundary detection by decision level
General approach
The decision-level boundary detection technique locates a boundary by finding a point in a data set which straddles a trigger intensity level, T T is determined from a baseline intensity,
I Base, a peak intensity, I Max and a fractional parameter, the decision factor K The boundary is then defined as the interpolation of two points, x L and x R , that straddle T
Figure C.1 illustrates a typical one-dimensional near-field intensity, where the cladding creates a shadow against a bright background, leaving the core un-illuminated The red line represents the baseline intensity level, while the blue line indicates the peak reference level, and the green line denotes the decision level, set at K of 0.5 (or 50%) The cladding intersects the threshold on both the left and right sides of the x-axis, with the fibre’s diameter defined as the distance between these two intersection points.
Figure C.1 – Typical one-dimensional data set, cladding only
Class A multimode fibre core reference level and k factor
Accurate estimation of reference levels is essential for determining the boundaries of a body, as these levels influence the decision level In cladding bodies, steep edge transitions mean that minor variations in the decision level have minimal impact on detected edges However, as shown in Figure C.2, the multimode core diameter is often defined using decision factors that identify features near the core boundary, where transitions are more gradual Consequently, small adjustments in reference levels can significantly alter the edge location, ultimately affecting the computed core diameter.
NOTE Right-hand graph is expanded in Y by 10
Figure C.2 – Typical graded index core profile
The upper reference level for the core of graded-index fibers is defined as the maximum intensity measurement within the core or an average of values near the peak For step-index multimode fiber cores, it is essential to determine the upper reference level similarly to the baseline, as the signal within the core may not be uniform Therefore, it is crucial to establish a reasonable upper reference level for these types of fibers.
In general, care should be taken to find repeatable and realistic baseline reference levels
In certain near-field transmission systems, such as those utilizing a modulated core illuminator and a demodulated signal, the baseline reference level is anticipated to be zero Conversely, for other systems, the baseline reference may differ from zero and must be established from the data set.
The default reference k-factor used for core diameter measurement of category A1 and A4 fibres is to be 0,025 (2,5 %), for category A2 and A3 fibres 0,5 (50 %) shall be used
All class A fibre specifications are currently under revision to incorporate the k-factor for core geometry determination Once the updated specifications are published, the previous information will be disregarded, and the new product specifications will take precedence.
For everyday measurements, alternative values of k from different core processing methods can be utilized In such instances, these non-reference measured values must be correlated with the reference value for k and the corresponding method, as outlined in Annex F.
Class B and C single-mode fibres
The core edge table for single-mode fibres primarily serves to identify the core center for calculating concentricity error, making the edge detection methodology less critical Utilizing the maximum pixel in the core region as the upper reference level is a reasonable approach For baseline reference level determination, refer to section C.2.2, keeping in mind that baseline level errors are typically less significant for these fibre classes A commonly used k factor is 0.25 (25%).
Direct geometry computation of one-dimensional data
After completing edge detection on a single-scan one-dimensional data set, the body's diameter can be calculated by finding the difference between the edges detected on the right and left sides of the scan.
To estimate concentricity, both the core and cladding bodies must be detected The center of each is calculated as the average of its left and right edges, and the concentricity estimate is determined by the difference between these two centers.
When scans are conducted at multiple angles, geometry can be calculated for each angle If three or more angular scans are obtained, it is advisable to compile their edges into an edge table, as outlined in Clause C.3, and to fit the data to an ellipse, as detailed in Annex D.
Assembling edge tables from raw data
General
An edge table is a collection of X,Y data pairs that represent the bounding points of a body, forming a nearly circular closed curve over 360° These tables are created from detected edges using specific methods and filters, and they are derived from two-dimensional raw intensity data, whether refracted or transmitted, as outlined in Annexes A and B Edge tables can be constructed from raster data sets or from multiple single scans captured at various angles.
Edge tables from raster data
Figure C.3 – Raster data, cladding only
Figure C.3 illustrates a standard raster scan from a video near-field instrument, featuring a dark cladding set against a bright background, with the core unlit To create a cladding edge table for this scenario, the image is analyzed on a pixel-by-pixel basis, focusing on the edges.
IEC detection techniques discussed in Clause C.2, and a list constructed of the locations (X,Y) of detected edges
In the analyzed image, each row and column can exhibit up to two detectable edges, while illuminated cores may reveal four edges in certain rows and columns Rows and columns outside the fiber area show no detectable edges The green line indicates a row close to the cladding's diameter, whereas the red line marks a row scanning nearly tangentially to the cladding Scans near the cladding's center yield the sharpest edges, while tangential scans result in weak, difficult-to-detect edges Therefore, it is crucial to focus on detecting edges in rows or columns that are as close to the center as possible.
An effective method for edge detection in images involves focusing on edge detection along the central rows, while employing column-wise edge detection for the surrounding periphery.
The optimal transition for detection occurs at 45° and 135° angles in the image, with the yellow line marking the point where the detection method should change from row-wise to column-wise.
An alternative method involves conducting edge detection solely on scans that traverse the rough center of the body To utilize the complete image, two-dimensional interpolation is applied to create synthetic one-dimensional scans at sufficiently fine angles to match the video resolution The chosen angular increment ensures that the arc length corresponds to the pixel spacing at the body's radius Subsequently, the detected edges from each synthetic scan are transformed into the zero-angle coordinate system and incorporated into the edge table.
When complete, a n e length table of X i ,Y i edges will be determined for each body analysed.
Edge tables from multi-angular one-dimensional scans
To assemble an edge table from a multi-angle scan set, process each scan as outlined in
In Clause C.2, it is crucial to reference the location of each detected edge to the fibre's rotation center Each identified body will have an associated list of \( n \) pairs of \( R_{k,\phi_k} \), where the \( R \) data represents detected edges from both the left and right sides of a scan The left side edges are indicated by negative values, as they are positioned to the left of the rotation center Subsequently, the \( R_{k,\phi_k} \) pairs should be converted into a Cartesian coordinate edge table using the transformations \( X = R_k \cdot \cos(\phi_k) \) and \( Y = R_k \cdot \sin(\phi_k) \).
Edge table ellipse fitting and filtering
Introductory remarks
The process of fitting an ellipse to an edge table involves applying specific parameters to both the core and cladding edge tables, which are then utilized to determine the geometry of the fiber.
General mathematical expressions for ellipse fitting
A general form for an ellipse is given as
0 = ax + 2 bxy cy + + 2 dx + 2 fy + g (D.2) where
1 2 cd bf x b ac af bd y b ac
The rotation of the ellipse, φ is given by
The major and minor radial dimensions of the ellipse are computed by
2 Major a c c a c b a ac b acg bdf gb cd
R af a c c a a b c ac b acg bdf gb cd
The ellipse can be expressed parametrically as
Major cos sin sin cos sin sin – cos cos y R
′ ϕ θ ϕ θ ϕ θ ϕ θ (D.6) or, in cylindrical coordinates as
To fit the pixel data one solves the following linear system:
Each summation above is computed using the n e data pairs of X,Y points in the edge table
Numerical precision in practical computers can significantly impact results, particularly when calculating small differences between large, similar numbers In the described system, precision issues primarily arise from using data pairs whose relative origin lies outside the fitting boundary For instance, if the cladding edge table's origin is set at the lower left corner of the image, both the x and y data sets will be positive To mitigate these errors, it is advisable to subtract a rough center located within the body from each x and y datum.
Edge table filtering
Active filtering involves the elimination of raw edge points that indicate cleave damage or other imperfections, such as dirt, from the collection of fitted edges An example of this edge filtering process is illustrated below.
After fitting the edges in the edge table, calculate the distance, \(d\), between each edge in the fitted set and the ellipse using Equation (C.8) If this distance exceeds \(T\) micrometres, remove the edge from the table and increase the count of rejected edges, \(N_{\text{bad}}\) Should \(N_{\text{bad}}\) surpass 1% of the total edges in the table, proceed to refit the model using only the remaining edges.
– Repeat the above steps until step c) is false.
Geometric parameter extraction
In this clause, the subscripts "cl" and "co" differentiate the elliptical fit parameters of the cladding and core bodies
Usinge the fitted ellipses, the following geometric parameters can be extracted:
X co ,Y co (àm): fitted core centre
R Major co (àm): major radius of the core
R Minor co (àm): minor radius of the core
Core diameter (àm): (R Major CO +R Minor CO )
Core non-circularity (%): 200 (R Major co – R Minor co )/Core diameter
X cl ,Y cl (àm): fitted cladding centre
R Major cl (àm): major radius of the cladding
R Minor cl (àm): minor radius of the cladding
Cladding diameter (àm): (R Major CL +R Minor CL )
Cladding non-circularity (%): 200 (R Major cl – R Minor cl )/Cladding diameter
Core/cladding concentricity error (àm): [(X cl − X co ) 2 + (Y cl −Y co ) 2 ] ẵ
Fitting category A1 core near-field data to a power law model
Introductory remarks
Annex E outlines the methodology for fitting a power law profile to raw near-field data from a category A1 fibre core, applicable to both transmitted and refracted data This technique allows for the determination of core diameter, core center (with certain limitations), and the power-law exponent, α Successful fitting typically necessitates pre-processing steps, which are detailed in Clause E.2, while Clause E.3 provides an in-depth description of the fitting methodology.
Preconditioning data for fitting
Motivation
The fitting process described in Clause E.3 requires a data set which satisfies two conditions: the data set is one-sided (only exists in positive radius) and, has a zero intensity baseline
(zero intensity outside the core region) Two-dimensional data from Annex A, raster scanning, and Annex B, grey-scale technique can be pre-processed in similar ways as described in
E.2.2 One-dimensional data from Method A or Method B share pre-processing requirements as described in E.2.3.
Transformation of a two-dimensional image to one-dimensional radial near-field
To convert a two-dimensional image of a category A1 fibre core into a one-dimensional data set, apply this processing method, which allows for fitting the data to the power law profile as outlined in the relevant literature.
Clause E.3 states that the images are generally gray-scale video images obtained through the transmitted near-field gray-scale method outlined in Annex B Additionally, raster images captured using the refracted near-field method from Annex A can also be processed using this technique.
The initial raster or image often contains extraneous areas beyond the core, such as surrounding cladding and the illumination field in gray-scale images These areas can introduce bias during the fitting process outlined in Clause E.3 To mitigate this, it is advisable to extract a square area around the core from the raw image for further analysis Since the baseline subtraction in Clause E.3 relies on data 1.2 times the nominal radial dimension of the core, focusing solely on this extracted area is recommended for optimal processing.
When an area of interest is extracted from the original image, the values of N Row and N Col will change This detail will be overlooked for the sake of brevity in the rest of this annex.
Using the image, the near-field centre is computed by finding the centre of gravity of each
To determine the centroid of an image, begin by identifying P Max and P Min, which represent the intensities of the brightest and dimmest valid pixels, respectively Next, calculate the threshold T based on these values.
Next, compute the following three summations over all pixels, excluding pixels with intensities less than T, over the row and column indices r and c:
Row Col Row Col Row Col
When P Min is significantly different from P Max, particularly when the cladding is illuminated, the centroid may be biased if the core image is not centered within the overall image To enhance centroid estimation in such scenarios, it is beneficial to subtract P Min or another estimate of the baseline from the image prior to calculating the centroid.
E.2.2.4 Computation of radial data functions
This computation step reduces the 2-D pixel data into a 1-D radial function by averaging the pixels in sets of nested and overlapping annular rings (centred on X 0 ,Y 0 ) of thickness 2W
(where W is 0,2 mm unless otherwise specified) centred on the optical centre of the fibre,
The rings are spaced at W micrometres, with the radial coordinate in the resulting data functions representing the radial centroid of the pixel coordinates within each ring, as defined in section E.2.2.3.
Figure E.1 illustrates the filtering concept, where the square grid represents the image's pixels Two concentric rings are centered on the optical center (X₀, Y₀): the outer ring is vertically hatched, while the inner ring is horizontally hatched Each ring has a width of 2W, with a W-wide overlapping region that is cross-hatched in the diagram The grayed pixels indicate those that will be averaged into the outer ring, as their centers lie within its boundary.
To compute the radial functions, first determine the maximum radius of a complete ring, which identifies the largest ring that can fit within the image without being truncated by its boundaries This involves calculating the shortest distance from the center of the image to its edge.
(E.4) where "min" finds the minimum of the four distances Next, compute the number of rings, N R , as
N R = D− (E.5) a) Allocate and zero the three summation arrays, S R (0 N R ), S I (0 N R ), and S N (0 N R )
For each and every pixel (on row r and column c), perform the following steps: b) Compute the radial coordinate:
W trunc R i (E.7) d) If i is less than or equal to N R then sum into both ring i and ring i-1
The overlapping-ring smoother is implemented through the above double sum Finally, for each ring, compute the parametric function pair by calculating the average radius and average intensity.
The camera's resolution and selected ring thickness can lead to some interior rings having no pixels, resulting in zero S N values In such instances, those rings should be omitted, causing subsequent array elements to shift up and decreasing N R Additionally, it is possible for two or more adjacent rings to share the same R̅ value.
In cases where the radii and intensities of adjacent rings are nearly identical (within 0.01 mm), it is necessary to average the radii and intensities of these rings This process involves replacing the adjacent rings with a single ring that has the averaged radius (\( \bar{R} \)) and averaged intensity, while also appropriately decrementing the count of rings (\( N_R \)).
Pre-processing of one-dimensional near-field data
One-dimensional near-field category A1 fibre core data can be obtained through various methods, including the refracted near-field method and the mechanical scanning transmitted near-field method This data typically exhibits both negative and positive radius intensity values To process this data, it is essential to identify the center where R = 0, as the fitting process can only utilize positive radii Once the center is established, radial positions can be adjusted accordingly The preferred approach for data processing is to fold the data around the center, reflecting the left side onto the right side, although extracting one side for individual processing is also an option.
Figure E.2 – Illustration of 1-D near-field preconditioning, typical video line
The input data are N pairs Rˊ i ,Iˊ i
The near-field center is determined by calculating the center of gravity of the measured radius profile from the image To locate the centroid, identify P Max and P Min, which represent the maximum and minimum intensities in the profile, and subsequently compute the threshold T.
Next, compute the following summations over the entire profile, excluding profile data with intensities less than T:
When P Min is significantly different from P Max, particularly when the cladding is illuminated, the centroid may be biased if the core image is not centered within the overall image To enhance centroid estimation in such scenarios, it is beneficial to subtract P Min or another estimate of the baseline from the image prior to calculating the centroid.
Once the centre is known, folding the profile is trivial:
The equation \$R = R' - R\$ (E.14) indicates that the vertical bars represent the absolute value After folding the data, it is beneficial to arrange the dataset in ascending order of \$R\$ to simplify the subsequent fitting algorithm.
Baseline subtraction
After calculating the radial functions, the N F ' function outside the core region will yield a baseline value, referred to as B This baseline can result from factors such as video dark signal, cladding illumination, or a non-zero cladding refractive index To effectively prepare the data for fitting, as outlined in Clause D.3, it is essential to subtract this baseline One method to determine B is by averaging N F ' over the radial range from 0.575 times the fiber's nominal core diameter.
0,6 times the nominal core diameter
There are cases where B is expected to be zero: for example, when a chop-in amplifier is used to demodulate a modulated signal from a one-dimensional mechanical near-field scan
In these cases it is allowable to take B as zero.
Fitting a power-law function to an category A1 fibre near-field profile
The conditioned near-field data from Clause E.2 is fit to the following power-law model:
The model described by equation (E.16) incorporates the maximum intensity, denoted as \$I_0\$, derived from the best-fit model, along with the power law shape factor \$\alpha\$ and the optimal core radius \$a\$ This model will be applied to the R and I data set, utilizing the least squares method to minimize the sum of squares, represented as \$S\$.
= ∑ − − (E.17) where i 10 and i 80 are the indices that bracket the data set where I lies between 10 % and
80 % of the maximum of I, respectively The reason to limit the fit region is two-fold: first, the
The 80% limit does not account for anomalies near the core center, while the 10% limit excludes the tail of these profiles, which deviate from the model due to diffusion and intentional design features.
To use Equation (E.17) as written, the data set should be established by in increasing R and ignore any data very near the core which falls below the 80 % limit
Minimizing S in Equation (E.17) necessitates the use of non-linear equation solving techniques, as the fit parameters I₀, α, and a are interdependent Conventional non-linear solvers often struggle to find a solution for specific data sets, highlighting the need for specialized methods To address this, Equation (E.16) can be reformulated by combining terms.
Equation (E.17) can be rewritten as
Combining the first two derivatives and solving simultaneously for I 0 and K, we get
From Equation (E.21) it can be observed that for any α, both K and I 0 can be calculated directly It is therefore possible to reduce the three-parameter nonlinear minimization of
Equation (E.17) to a one-parameter minimization of Equation (E.19) by exploiting
To solve the system, we utilize Equation (E.18) with a one-dimensional nonlinear solver, specifically Newton’s method, applied to α The kernel function is computed using Equation (E.21) to determine K and I₀, while Equation (E.19) serves as the function to be minimized.
Once the solution is found, the core diameter is found as twice a, which is computed from K, using Equation (E.18)
Mapping class A core diameter measurements
Introductory remarks
Annex B, along with Annexes C and D, outlines the reference test method (RTM) for determining the core diameter of class A multimode fibre The sample lengths for various categories of A fibre can reach hundreds of metres, as specified in the detail specification However, for practical day-to-day measurements, it is preferable to use shorter lengths, such as 2 metres, to determine the core diameter without the need for stress-free deployment of long fibre lengths.
The methodology outlined in Annex C for determining the core boundary delineation curve may be impractical when short fiber lengths are used under overfilling launch conditions To address these challenges, it is advisable to adapt the mapping of the reference test conditions to more practical testing scenarios.
If alternate measurement conditions are employed for daily production measurements, the alternate condition’s core diameter can be transformed to estimate the reference condition diameter.
Mapping function
To establish a steady-state bias in fibre measurement, it is essential to compare the reference test method for class A fibre core diameter with an alternative method that may utilize a shorter test length or different analysis techniques If a bias is confirmed, a mapping function can be applied to convert the core diameter obtained from the alternative method into an approximation of the diameter from the reference method These mapped diameters can be reported as the core diameter, and the mapping function can be designed in various forms.
Or any other provably utile function, f:
IEC 60793-1-45, Optical fibres – Part 1-45: Measurement methods and test procedures –
4 Vue d’ensemble de la méthode 54
4.2.2 Sources d'erreur du balayage unidimensionnel 55
4.3.1 Combinaison simple de jeux de balayages à quelques angles 57
4.3.2 Ajustement d'ellipse de séries de données à plusieurs angles ou rémanentes 57
11 Informations à mentionner dans la spécification 58
Annexe A (normative) Exigences spécifiques à la Méthode A – Champ proche réfracté 59
A.2.4 Positionneur XYZ (table de balayage) 60
A.2.6 Optique de collecte et détecteur 61
A.4.1 Chargement et centrage de la fibre 62
A.5 Calcul de l'indice de réfraction 63
Annexe B (normative) Exigences spécifiques à la Méthode B – Champ proche transmis 66
B.2.3 Support de fibre et appareillage de positionnement 68
B.2.4 Extracteur de modes de gaine 68
B.2.7 Moniteur d'image vidéo (technique de la vidéo en niveaux de gris) 70
Annexe C (normative) Détection des limites et construction d'un tableau représentatif des limites 74
C.2 Détection de limite par la méthode du niveau de décision 74
C.2.2 Niveau de référence du cœur de fibre multimodale de classe A et facteur k 75 C.2.3 Fibres unimodales de classes B et C 76
C.2.4 Calcul de géométrie directe des données unidimensionnelles 77
C.3 Constitution de tableaux représentatifs de limite à partir des données brutes 77
This section discusses the creation of representative tables of limits based on residual data and the development of these tables from multi-angular one-dimensional scans Additionally, it includes a normative annex on elliptical adjustment and filtering of the representative limit tables.
D.2 Expressions mathématiques générales relatif à l'ajustement elliptique 80
D.3 Filtrage du tableau représentatif des limites 81
Annexe E (informative) Ajustement des données en champ proche du cœur de catégorie A1 à un modèle de loi de puissance 83
E.2 Préconditionnement des données pour l'ajustement 83
E.2.2 Transformation d'une image bidimensionnelle en un champ proche radial unidimensionnel 83 E.2.3 Prétraitement des données en champ proche unidimensionnelles 86
E.3 Ajustement d'une fonction de loi de puissance à un profil en champ proche de fibre de catégorie A1 88
Annexe F (informative) Correspondances des mesures de diamètre de cœur de classe
Figure 1 – Echantillonnage sur une corde 55
Figure 2 – Balayage d'un corps non-circulaire 56
Figure A.1 – Méthode du champ proche réfracté – Cellule 60
Figure A.2 – Montage des instruments type 60
Figure A.3 – Balayage de ligne de profil d'indice type dans une fibre de catégorie A1 63
Figure A.4 – Profil d'indice rémanent type sur une fibre de catégorie A1 64
Figure B.1 − Montage type, technique des niveaux de gris 67
Figure B.2 – Montage type, technique du balayage mécanique 67
Figure B.3 – Balayage en champ proche unidimensionnel type, cœur de catégorie A1 72
Figure B.4 – Données en champ proche rémanentes types, fibre de catégorie A1 73
Figure C.1 – Série de données unidimensionnelles type, relative à la gaine seulement 75
Figure C.2 – Profil de cœur à gradient d'indice type 76
Figure C.3 – Données rémanentes, gaine uniquement 78
Figure E.2 – Illustration du préconditionnement en champ proche unidimensionnel, ligne vidéo type 87
Partie 1-20: Méthodes de mesure et procédures d'essai –
The International Electrotechnical Commission (IEC) is a global standards organization comprising national electrotechnical committees Its primary goal is to promote international cooperation on standardization issues in the fields of electricity and electronics To achieve this, the IEC publishes international standards, technical specifications, technical reports, publicly accessible specifications (PAS), and guides, collectively referred to as "IEC Publications." The development of these publications is entrusted to study committees, which allow participation from any national committee interested in the subject matter Additionally, international, governmental, and non-governmental organizations collaborate with the IEC in its work The IEC also works closely with the International Organization for Standardization (ISO) under conditions established by an agreement between the two organizations.
The official decisions or agreements of the IEC on technical matters aim to establish an international consensus on the studied topics, as the relevant national committees of the IEC are represented in each study committee.
The IEC publications are issued as international recommendations and are approved by the national committees of the IEC While the IEC makes every reasonable effort to ensure the technical accuracy of its publications, it cannot be held responsible for any misuse or misinterpretation by end users.
To promote international uniformity, IEC National Committees strive to transparently implement IEC Publications in their national and regional documents Any discrepancies between IEC Publications and corresponding national or regional publications must be clearly stated in the latter.
The IEC does not issue any conformity certificates itself Instead, independent certification bodies offer compliance assessment services and, in certain sectors, utilize IEC conformity marks The IEC is not responsible for any services provided by these independent certification organizations.
6) Tous les utilisateurs doivent s'assurer qu'ils sont en possession de la dernière édition de cette publication
The IEC, along with its directors, employees, agents, and committee members, shall not be held liable for any injuries, damages, or losses of any kind, whether direct or indirect, arising from the publication or use of this IEC Publication or any other IEC Publication, including any associated costs or legal expenses.
8) L'attention est attirée sur les références normatives citées dans cette publication L'utilisation de publications référencées est obligatoire pour une application correcte de la présente publication
Attention is drawn to the fact that some elements of this IEC publication may be subject to patent rights The IEC cannot be held responsible for failing to identify such patent rights or for not reporting their existence.
La Norme internationale IEC 60793-1-20 a été établie par le sous-comité SC86A: Fibres et câbles du comité d'études TC86 de l'IEC: Fibres optiques
Cette deuxième édition annule et remplace la première édition, publiée en 2001, dont elle constitue une révision technique
Cette édition contient les modifications techniques principales suivantes par rapport à l'édition précédente:
The reference testing method for all types of fibers has been updated from the refracted near-field method to the transmitted near-field method using grayscale video.
• les longueurs d'essai de tous les types de fibres doivent désormais être spécifiées dans la spécification particulière de la fibre;
Although default values are provided, the illumination wavelength of the core for all types of multimode fibers can now be specified in the fiber's particular specification.
• le facteur k appliqué sur le cœur (niveau de décision) doit maintenant être spécifié dans la spécification particulière pour tous les types de fibres multimodales;
• la présente édition décrit le mesurage de manière bien plus spécifique; la réduction et la transformation des données sont décrites de manière exhaustive
• la méthodologie de réduction des données pour la méthode du champ proche réfracté et pour la méthode du champ proche transmis est maintenant unifiée et cohérente
Le texte de cette norme est issu des documents suivants:
Le rapport de vote indiqué dans le tableau ci-dessus donne toute information sur le vote ayant abouti à l'approbation de cette norme
Cette publication a été rédigée selon les Directives ISO/IEC, Partie 2
Une liste de toutes les parties de la série IEC 60793, publiées sous le titre général Fibres optiques, est disponible sur le site internet de l’IEC
The committee has determined that the content of this publication will remain unchanged until the stability date specified on the IEC website at "http://webstore.iec.ch" for the relevant publication data On that date, the publication will be updated.
• remplacée par une édition révisée, ou
IMPORTANT – The "colour inside" logo on the cover of this publication indicates that it contains colors essential for a better understanding of its content Therefore, users should print this document using a color printer.
La présente norme présente deux méthodes destinées à mesurer les caractéristiques géométriques de la fibre:
– Méthode A: Champ proche réfracté, décrite dans l'Annexe A;
– Méthode B: Champ proche transmis, décrite dans l'Annexe B
Methods A and B are used to measure the geometry of all multimodal fibers of class A, unimodal fibers of class B, and unimodal interconnection fibers of class C The applicable product specifications for these fibers are outlined in IEC 60793-2.
10, l’IEC 60793-2-20, l’IEC 60793-2-30, l’IEC 60793-2-40, l’IEC 60793-2-50 et l'IEC 60793-2-
60, fournissent les détails de mesure correspondants, y compris les longueurs d’échantillons et les facteurs k
Les paramètres géométriques mesurables par les méthodes décrites dans la présente norme sont les suivants:
– diamètre du cœur (fibre de catégorie A seulement);
– non-circularité du cœur (fibre de catégorie A seulement);
– erreur de concentricité entre le cœur et le gainage
NOTE 1 Le diamètre de cœur des fibres de classe B et de classe C n'est pas spécifié Le paramètre équivalent est le diamètre de champ de mode, déterminé par l'IEC 60793-1-45
These methods outline the techniques for data collection and analysis in both unidimensional (1-D) and bidimensional (2-D) formats However, 1-D methods alone are insufficient for assessing non-circularity or concentricity errors When measuring non-circular bodies, the limitations of these methods become evident.
1-D, les diamètres des corps présentent des incertitudes supplémentaires Ces limitations peuvent être résolues par le balayage et l'analyse de plusieurs séries de données 1-D L'Article 5 donne de plus amples informations
Les informations communes aux deux méthodes apparaissent dans les Articles 2 à 10, et les informations spécifiques à chaque méthode apparaissent, respectivement, dans les Annexes