EN 4533-002:2006

2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.PHUMHQMHLuft- und Raumfahrt - Faseroptische Systemtechnik - Handbuch - Teil 002: Prüfung und MessungSérie aérospatiale - Systèmes des fibres optiques - Manuel d'utilisation - Partie 002 : Essais et mesuresAerospace series - Fibre optic systems - Handbook - Part 002: Test and measurement49.060Aerospace electric equipment and systemsICS:Ta slovenski standard je istoveten z:EN 4533-002:2006SIST EN 4533-002:2009en,de01-junij-2009SIST EN 4533-002:2009SLOVENSKI
STANDARD
EUROPEAN STANDARDNORME EUROPÉENNEEUROPÄISCHE NORMEN 4533-002July 2006ICS 49.060 English VersionAerospace series - Fibre optic systems - Handbook - Part 002:Test and measurementSérie aérospatiale - Systèmes des fibres optiques - Manueld'utilisation - Partie 002 : Essais et mesuresLuft- und Raumfahrt - Faseroptische Systemtechnik -Handbuch - Teil 002: Tests und MessungenThis European Standard was approved by CEN on 28 April 2006.CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this EuropeanStandard the status of a national standard without any alteration. Up-to-date lists and bibliographical references concerning such nationalstandards may be obtained on application to the Central Secretariat or to any CEN member.This European Standard exists in three official versions (English, French, German). A version in any other language made by translationunder the responsibility of a CEN member into its own language and notified to the Central Secretariat has the same status as the officialversions.CEN members are the national standards bodies of Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France,Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania,Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.EUROPEAN COMMITTEE FOR STANDARDIZATIONCOMITÉ EUROPÉEN DE NORMALISATIONEUROPÄISCHES KOMITEE FÜR NORMUNGManagement Centre: rue de Stassart, 36
B-1050 Brussels© 2006 CENAll rights of exploitation in any form and by any means reservedworldwide for CEN national Members.Ref. No. EN 4533-002:2006: ESIST EN 4533-002:2009

 Part 003: Looming and installation practices  Part 004: Repair, maintenance and inspection b) Background It is widely accepted in the aerospace industry that photonic technology offers a number of significant advantages over conventional electrical hardware. These include massive signal bandwidth capacity, electrical safety, and immunity of passive fibre-optic components to the problems associated with electromagnetic interference (EMI). To date, the latter has been the critical driver for airborne fibre-optic communications systems because of the growing use of non-metallic aerostructures. However, future avionic requirements are driving bandwidth specifications from 10’s of Mbits/s into the multi-Gbits/s regime in some cases, i.e. beyond the limits of electrical interconnect technology. The properties of photonic technology can potentially be exploited to advantage in many avionic applications, such as video/sensor multiplexing, flight control signalling, electronic warfare, and entertainment systems, as well as in sensing many of the physical phenomena on-board aircraft. The basic optical interconnect fabric or `optical harness’ is the key enabler for the successful introduction of optical technology onto commercial and military aircraft. Compared to the mature telecommunications applications, an aircraft fibre-optic system needs to operate in a hostile environment (e.g. temperature extremes, humidity, vibrations, and contamination) and accommodate additional physical restrictions imposed by the airframe (e.g. harness attachments, tight bend radii requirements, and bulkhead connections). Until recently, optical harnessing technology and associated practices were insufficiently developed to be applied without large safety margins. In addition, the international standards did not adequately cover many aspects of the life cycle. The lack of accepted standards thus lead to airframe specific hardware and support. These factors collectively carried a significant cost penalty (procurement and through-life costs), that often made an optical harness less competitive than an electrical equivalent.
c) The fibre-optic harness study The Fibre-Optic Harness Study concentrated on developing techniques, guidelines, and standards associated with the through-life support of current generation fibre-optic harnesses applied in civil and military airframes (fixed and rotary wing). Some aspects of optical system design were also investigated. This programme has been largely successful. Guidelines and standards based primarily on harness study work are beginning to emerge through a number of standards bodies. Because of the aspects covered in the handbook, European prime contractors are in a much better position to utilise and support available fibre optic technology. SIST EN 4533-002:2009

This Part of EN 4533 will explain the measurement problem and the techniques used to overcome them in greater detail. 2 Normative references The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. EN 4533-004,
Aerospace series – Fibre optic systems – Handbook – Part 004: Repair, maintenance and inspection.
ARP5061,
Guidelines for Testing and Support of Aerospace, Fiber Optic, Inter-Connect Systems. 1) 3 Problem areas and limitations 3.1 The problem of testing avionic, multi-mode fibre installations The insertion loss of a fibre optic harness can be divided into two contributions. The first is the intrinsic loss of the harness caused by the properties of the materials such as the absorption of the silica of the fibre core. In the case of multi-mode fibres, this would include the variation in the loss of the component caused by changes in the power distribution. When discussing insertion loss, the term ‘power distribution’ will be used to describe the spatial and angular variation of the power across the fibre’s core rather than the temporal variation of the power along the length of the fibre. The second contribution is the additional loss that is introduced into the harness from extrinsic losses such as misalignment errors in connectors and contamination.
The insertion loss of components used in any fibre optic link depends on the power distribution of the light that passes through them. However, in some types of fibre harness, the shape of the power distribution does not change as the light propagates through it and the component insertion loss is independent of its position within the harness. For example, the power distribution in single-mode fibre harnesses is fixed by the fibre parameters and the source’s wavelength. Long-haul (a few kilometres between components), multi-mode fibre harnesses also effectively have a fixed power distribution because the distance between components is sufficient for the power distribution to reach an equilibrium state that depends only on the fibre parameters. The insertion loss of a long-haul multi-mode harness component is defined by this equilibrium power distribution.
 1) Published by: Society of Automotive Engineers (SAE), 400 Commonwealth Drive, Warrendale, PA 15096-0001. SIST EN 4533-002:2009

a) A large area detector collects light from the fibre that is outside of the defined limits for the source. b) The removal of an interface by not using a test lead between the lead under test and the power meter. Figure 1 — Causes of unrepresentative power measurements
3.2.4 Optical time domain reflectometry Optical time domain reflectometry (OTDR) is a single ended diagnostic/measurement technique that relies on the backscatter of light from ‘imperfections’ and discontinuities in a fibre-optical system. It is used extensively in the telecommunication industry for optical system commissioning and testing. OTDR technology potentially enables a reference insertion loss footprint to be generated requiring only a single measurement to characterise an entire harness under test. Furthermore, comparison between current and previous traces can be performed automatically by standard OTDR software largely de-skilling measurement and diagnostic operations. However, current OTDRs struggle to meet the specific requirements of airborne optical harnesses, in particular spatial resolution (dead zone) performance. 3.3 The way forward The following three options would enable a more reliable means of predicting and testing the performance of multi-mode fibre optic harnesses: 1) A method of predicting the power distribution throughout the harness. Those power distributions can be used to find an exact value for the insertion loss of each component. The insertion loss will then be specific to that particular component at that particular position within the harness. SIST EN 4533-002:2009

Ideally, component manufacturers would not specify a loss of a component specifically but define how the input launch is changed by the component before launching into the next component by means of a component matrix. This matrix method will be described in 5.4. Using this approach, a representative loss of the component can be achieved that is accurate for all launch conditions. The problem with this method is that, at present, there is no internationally accepted practical test for component manufacturers to determine the matrix elements.
A more practical approach is for the component manufacturer to specify an optical performance figure based on a test that is representative for short haul applications. These test conditions have been defined using raytracing software developed within the Fibre Optic Harness Study and these may require different test sources for the specific fibre types being used. If the structure and typical tolerances of a commercial component are known, it should be possible to use the raytracing model that will be described in the next section to convert the insertion loss attained with the manufacturer’s launch condition into the loss for a standard, avionic launch condition. 4.3 Computer modelling A computer can be used to predict the way in which the optical power distribution in an optical system changes as it passes through components. There are many optical design packages in the marketplace which can be roughly divided into two types. The first are lens design packages that use raytracing to predict the power distributions in lens systems such as those found in photographic equipment. The second are waveguide design packages that calculate the strength of the light’s electric field to find the power distributions in integrated optic components. Raytracing is a valid technique where it can be assumed that the apertures of the components are large compared to the wavelength of the light passing through them. Multi-mode fibres have core diameters that normally fulfil this criterion and it is therefore valid to use raytracing to predict the power distributions in multi-mode harnesses. During the Fibre Optic Harness Study, a multi-mode fibre optic system design package was written that uses raytracing to predict the power distributions. This model was validated against experimental data. Figure 2 is a view of the model screen showing the various components that can be connected together to construct a complete harness. The model can be used to predict the power distribution at any point in the harness and can calculate the loss of individual components. Amongst other facilities, typical component errors can be introduced, as can representations of contamination. It is also capable of calculating the component matrices that will be described in the following section.  2) For graded index fibre the best correlation between the two estimates of system loss was found for a useable power definition of 95 % of the core diameter and 95 % of the maximum acceptance angle. In step index fibre the best definition of useable power was found to be 100 % of the core diameter and 90 % of the maximum acceptance angle. SIST EN 4533-002:2009

Figure 2 — The user interface for the raytracing model showing the various components that can be included 4.4 Matrices In a fibre system in which the power distribution does not change, it is possible to represent the component losses by a single number, the insertion loss. As discussed in Clause 3, this applies to single-mode fibre systems and multi-mode harnesses where the component separation is large, e.g. telecommunication links. In shorter links, a single insertion loss value will say nothing about the way in which the component’s loss is influenced by the power distribution into it or how the component itself alters the power distribution. Representing the components as matrices can overcome these problems. [2][3][4]
Key 1 Input light 2 Component 3 Output light Figure 3 — Matrix representation of a fibre-optic component 21 3A’ B’ C’ A B C
l11SIST EN 4533-002:2009
In a complex harness with a number of components, a system matrix can be created by multiplying all of the individual component matrices together. Matrix multiplication is different to ordinary multiplication. If two 3 by 3 matrices, M and N, had to be multiplied together that had elements: ==333231232221131211333231232221131211
and
nnnnnnnnnNmmmmmmmmmM then the resulting matrix, R, would be: ++++++++++++++++++==333323321331323322321231313321321131332323221321322322221221312321221121331323121311321322121211311321121111.nmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmnmNMR The order in which the components are multiplied together is crucial. If a system is comprised of components with matrices, A B C and D, and they appeared in this order going from the source to the receiver, the matrix for the complete system, R, will be the multiple of the matrices in the reverse order: ABCDR.= This is further illustrated in Figure 4. SIST EN 4533-002:2009

Key 1 Source 2 Receiver 3 System matrix R = D.C.B.A Figure 4 — The order in which component matrices are multiplied to find the matrix of the system In this way, a system designer can use these matrices to construct a single matrix that describes how the system will alter and attenuate the light passing through it. By multiplying this matrix by the column vector that represents the input power distribution, the designer can predict the insertion loss of a system and the output power distribution. The number of elements in the matrix is rather arbitrary. The larger the number, the greater the accuracy of representing the power distribution. In the past, 2 by 2 matrices (4 elements) have been used but the accuracy has not been very good [5]. A better performance has been obtained with 3 by 3 matrices [6]. The raytracing model can generate matrices with up to 25 elements (5 by 5 matrix). The raytracing model can generate matrices by tracking the variation of the parameters of rays as they pass through the harness components. However, it is much more difficult to extract the matrices from practical measurements of the power distribution. This is particularly true of harness components used with graded-index fibre because the parameter used to generate the matrix is not directly related to either the spatial (near-field) or angular (far-field) power distributions [7]. Matrices for graded-index fibre harnesses have been successfully generated from near-field power distributions [8]. Compared to graded-index fibre components, matrices for components in a step-index fibre harness are relatively easy to generate from the far-field power distribution. The technique can be illustrated by Figure 3. An optical system launches light at angles within region A into the component. A power meter measures the power within output regions A’, B’ and C’. The values of the matrix elements m11, m12 and m13 are: ACmABmAAm inpower ' inpower
inpower ' inpower
inpower ' inpower 131211=== The other matrix elements are found by launching light into regions B or C and measuring the power in the three output regions. The matrix elements are: BCmBBmBAm inpower ' inpower
inpower ' inpower
inpower ' inpower 232221=== CCmCBmCAm inpower ' inpower
inpower ' inpower
inpower ' inpower 333231=== A C1 2 BD3SIST EN 4533-002:2009

An additional parameter that affects the launch condition is whether the fibre core has a step or graded refractive index profile. This is very important because the shape of the near and far-field distributions is strongly dependent on the index profile, as illustrated in Figure 5. ) and b) are the near and far-field power distributions from a graded-index fibre illuminated by an LED source. The LED source provides a nearly fully-filled launch condition. c) and d) are the near and far-field power distributions of a step-index fibre that is illuminated by a white light source. The source provides an overfilled launch condition.
a) 0.26 NA graded-index fibre with an LED source near-field b) 0.26 NA graded-index fibre with an LED source far-field
c) 0.2 NA step index fibre with white light source near-field d) 0.2 NA step index fibre with white light source far-field
Figure 5 The launch condition is expressed as a percentage of the core radius and numerical aperture of the fibre in the harness. Say, for example, that a harness uses a fibre with a core radius of 50 µm and a numerical aperture of 0,29. If the near-field profile of the source corresponds to a fibre with a radius of 40 µm and the far-field to a numerical aperture of 0,25, the launch condition into the fibre will be 80:86 (80 % fill of the core radius and an 86 % fill of the numerical aperture). If the launch condition of the source is 100:100, this is called a ‘fully filled’ launch. If either of the percentages are greater than 100, the launch is ‘overfilled’ in either the near or far-field distribution. If either of the percentages are less than 100, the launch is ‘under filled’ in the corresponding distribution. One of the important implications of the launch condition being defined by the fibre parameters is that the launch condition of a source will change depending on the fibre type and parameters being used in the harness. Thus two different sources will be needed to produce identical launch conditions in two fibres with different parameters.
5.2.4 Why do we need to condition the test source? As discussed in the introduction, making measurements of the insertion loss of components used in multi-mode optical fibre harnesses is difficult. It is complicated by the fact that the loss depends on the power distribution of the source that is used to make the measurement. It is possible to obtain a very large range of loss figures for a system by using different test sources. For example, a source that injects most of its power into a very narrow angle inside the component will measure a much smaller loss compared to a source that injects power over a larger range of angles. This is because light at the higher angles is more likely to be attenuated by imperfections in the construction of the component than light at smaller angles. Position (microns) Position (µm) 20010 20 30 40 50 0 20 40 60 80100 120 140 160 180Power (nW) 123456-20-15-10-50510 15 20 Angle of exit (degrees) Power (nW) 0 -50 50 4
6 010 -100.20.40.6Angle (degrees) Power (nW) Power (nW) SIST EN 4533-002:2009

1) The light falling on the detector has to be conditioned so that only light within the defined standard power distribution falls on it. 2) The detector dimensions have to be larger than the area of the patch of light that falls on it (see 5.3). 5.2.6 Optimum launch conditions One solution to this measurement problem is for everyone to agree to always use a standard power distribution. But which distribution is the correct one? The answer is the distribution that is most characteristic of the harness system being measured. For example, telecommunication harnesses have long lengths of fibre between components and in these distances it is possible for the light power distribution to reach an equilibrium condition. This means that the power distribution no longer changes shape as it propagates down the fibre but simply reduces in power. For telecommunication fibre, this equilibrium power distribution is obviously the correct power distribution to use. Unfortunately, it is more difficult to define the power distribution for short-haul avionic harnesses.
In avionic harnesses, a distribution that fully fills the fibre core and the numerical aperture will lose light launched close to the core boundary and the maximum angle allowed by the fibre’s numerical aperture. The resulting distribution will slightly underfill the fibre and is the distribution that should be used to make insertion loss measurements. The same distribution will eventually be obtained by a very ‘underfilled’ launch in a harness with a large number of components because the manufacturing imperfections in the components will scatter light into higher angles and core positions. This has been developed in the United States as a practical technique to find the optimum distribution for an avionic harness [10]. A test harness with typical components is constructed and the optimum distribution found by making measurements of the output power distribution as the input distribution is altered. This has resulted in limits being set to the launch condition of the Spectran 'Flightguide' [11]. The limits for this fibre have been set at three percentage levels (75 %, 15 % and 5 %) within the normalized near and far-field power distributions. These limits are shown in Figure 6, together with the curve of ‘ideal’ near and far-field power distributions for a fibre with 'Flightguide' fibre parameters. The ‘ideal’ distributions fall quadratically from the maximum intensity to the value of the core radius or numerical aperture. Test equipment with these launch conditions is available as
...