EN 60793-1-33:2002

STANDARDOptična vlakna – 1-33. del: Metode merjenja in preskusni postopki – Dovzetnost na napetostno korozijo (IEC 60793-1-33:2001)*Optical fibres - Part 1-33: Measurement methods and test procedures - Stress corrosion susceptibility (IEC 60793-1-33:2001)©
Standard je založil in izdal Slovenski inštitut za standardizacijo. Razmnoževanje ali kopiranje celote ali delov tega dokumenta ni dovoljenoReferenčna številkaSIST EN 60793-1-33:2004(en)ICS33.180.10

EUROPEAN STANDARDEN 60793-1-33NORME EUROPÉENNEEUROPÄISCHE NORMApril 2002CENELECEuropean Committee for Electrotechnical StandardizationComité Européen de Normalisation ElectrotechniqueEuropäisches Komitee für Elektrotechnische NormungCentral Secretariat: rue de Stassart 35, B - 1050 Brussels© 2002 CENELEC -All rights of exploitation in any form and by any means reserved worldwide for CENELEC members.Ref. No. EN 60793-1-33:2002 EICS 33.180.10English versionOptical fibresPart 1-33: Measurement methods and test procedures -Stress corrosion susceptibility(IEC 60793-1-33:2001)Fibres optiquesPartie 1-33: Méthodes de mesureet procédures d'essai -Résistance à la corrosion sous contrainte(CEI 60793-1-33:2001)LichtwellenleiterTeil 1-33: Messmethodenund Prüfverfahren -Spannungskorrosionsempfindlichkeit(IEC 60793-1-33:2001)This European Standard was approved by CENELEC on 2002-03-05. CENELEC members are bound tocomply 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 national standards may be obtained onapplication to the Central Secretariat or to any CENELEC member.This European Standard exists in three official versions (English, French, German). A version in any otherlanguage made by translation under the responsibility of a CENELEC member into its own language andnotified to the Central Secretariat has the same status as the official versions.CENELEC members are the national electrotechnical committees of Austria, Belgium, Czech Republic,Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Malta, Netherlands,Norway, Portugal, Spain, Sweden, Switzerland and United Kingdom.

- 3 -EN 60793-1-33:2002Annex ZA(normative)Normative references to international publicationswith their corresponding European publicationsThis European Standard incorporates by dated or undated reference, provisions from otherpublications. These normative references are cited at the appropriate places in the text and thepublications are listed hereafter. For dated references, subsequent amendments to or revisions of anyof these publications apply to this European Standard only when incorporated in it by amendment orrevision. For undated references the latest edition of the publication referred to applies (includingamendments).NOTEWhen an international publication has been modified by common modifications, indicated by (mod), the relevantEN/HD applies.PublicationYearTitleEN/HDYearIEC/TR 62048- 1)The law theory of optical fibre reliability--
1) To be published.
NORMEINTERNATIONALECEIIECINTERNATIONALSTANDARD60793-1-33Première éditionFirst edition2001-08Fibres optiques –Partie 1-33:Méthodes de mesures et procédures d'essai –Résistance à la corrosion sous contrainteOptical fibres –Part 1-33:Measurement methods and test procedures –Stress corrosion susceptibility Commission Electrotechnique Internationale International Electrotechnical CommissionPour prix, voir catalogue en vigueurFor price, see current catalogue IEC 2001
Droits de reproduction réservés

Copyright - all rights reservedAucune partie de cette publication ne peut être reproduite niutilisée sous quelque forme que ce soit et par aucun procédé,électronique ou mécanique, y compris la photocopie et lesmicrofilms, sans l'accord écrit de l'éditeur.No part of this publication may be reproduced or utilized inany form or by any means, electronic or mechanical,including photocopying and microfilm, without permission inwriting from the publisher.International Electrotechnical Commission3, rue de Varembé
Geneva, SwitzerlandTelefax: +41 22 919 0300e-mail: inmail@iec.ch IEC web site
http://www.iec.chCODE PRIXPRICE CODEW

60793-1-33 © IEC:2001– 3 –CONTENTSFOREWORD.5INTRODUCTION.91Scope and object.112Normative references.113Apparatus.134Sampling and specimens.135Reference test method.136Procedure.157Calculations.158Results.159Specification information.15Annex A (normative)
Dynamic n value by axial tension.17Annex B (normative)
Dynamic n value by two-point bending.31Annex C (normative)
Static n value by axial tension.41Annex D (normative)
Static n value by two-point bending.47Annex E (normative)
Static n value by uniform bending.51Annex F (informative)
Considerations for dynamic fatigue calculations.57Annex G (informative)
Considerations for static fatigue calculations.65Annex H (informative)
Considerations on stress corrosion susceptibility parametertest methods.67Bibliography.75Figure A.1 – Schematic of translation test apparatus.17Figure A.2 – Schematic of rotational test apparatus.19Figure A.3 – Schematic of rotational test apparatus with load cell.19Figure A.4 – Representation of dynamic fatigue graph.29Figure B.1 – Schematic of two-point bending unit.37Figure B.2 – Schematic of surface platen.39Figure B.3 – Dynamic fatigue data schematic.39Figure C.1 – Schematic of possible static fatigue (tension) apparatus.45Figure D.1 – Schematic of possible static fatigue (two-point bending) apparatus .49Figure E.1 – Schematic of possible static fatigue (uniform bending) apparatus.55Figure H.1 – The results of the round robin fracture strength versus time.73Figure H.2 – The results of the round robin fracture strength versus time.73Table F.1 − 95 % confidence interval for nd.59

60793-1-33 © IEC:2001– 5 –INTERNATIONAL ELECTROTECHNICAL COMMISSION____________OPTICAL FIBRES –Part 1-33: Measurement methods and test procedures –Stress corrosion susceptibilityFOREWORD1)The IEC (International Electrotechnical Commission) is a worldwide organization for standardization comprisingall national electrotechnical committees (IEC National Committees). The object of the IEC is to promoteinternational co-operation on all questions concerning standardization in the electrical and electronic fields. Tothis end and in addition to other activities, the IEC publishes International Standards. Their preparation isentrusted to technical committees; any IEC National Committee interested in the subject dealt with mayparticipate in this preparatory work. International, governmental and non-governmental organizations liaisingwith the IEC also participate in this preparation. The IEC collaborates closely with the InternationalOrganization for Standardization (ISO) in accordance with conditions determined by agreement between thetwo organizations.2)The formal decisions or agreements of the IEC on technical matters express, as nearly as possible, aninternational consensus of opinion on the relevant subjects since each technical committee has representationfrom all interested National Committees.3)The documents produced have the form of recommendations for international use and are published in the formof standards, technical specifications, technical reports or guides and they are accepted by the NationalCommittees in that sense.4)In order to promote international unification, IEC National Committees undertake to apply IEC InternationalStandards transparently to the maximum extent possible in their national and regional standards. Anydivergence between the IEC Standard and the corresponding national or regional standard shall be clearlyindicated in the latter.4)
The IEC provides no marking procedure to indicate its approval and cannot be rendered responsible for anyequipment declared to be in conformity with one of its standards.5)
Attention is drawn to the possibility that some of the elements of this International Standard may be the subjectof patent rights. The IEC shall not be held responsible for identifying any or all such patent rights.International Standard IEC 60793-1-33 has been prepared by subcommittee 86A: Fibres andcables, of IEC technical committee 86: Fibre optics.This standard, together with the other standards in the IEC 60793-1-3X series, cancels andreplaces the second edition of IEC 60793-1-3, of which it constitutes a technical revision.The text of this standard is based on the following documents:FDISReport on voting86A/688/FDIS86A/727/RVDFull information on the voting for the approval of this standard can be found in the report onvoting indicated in the above table.This publication has been drafted in accordance with the ISO/IEC Directives, Part 2.Annexes A, B, C, D, E form an integral part of this standard.Annexes F, G, H are for information only.

60793-1-33 © IEC:2001– 7 –IEC 60793-1-3X consists of the following parts, under the general title Optical fibres:• Part 1-30: Measurement methods and test procedures: Fibre proof test• Part 1-31: Measurement methods and test procedures: Tensile strength• Part 1-32: Measurement methods and test procedures: Coating strippability• Part 1-33: Measurement methods and test procedures: Stress corrosion susceptibility• Part 1-34: Measurement methods and test procedures: Fibre curlThe committee has decided that the contents of this publication will remain unchangeduntil 2003. At this date, the publication will be•reconfirmed;•withdrawn;•replaced by a revised edition, or•amended.

60793-1-33 © IEC:2001– 9 –INTRODUCTIONPublications in the IEC 60793-1 series concern measurement methods and test procedures asthey apply to optical fibres.Within the same series several different areas are grouped, as follows:–parts 1-10 to 1-19: General–parts 1-20 to 1-29: Measurement methods and test procedures for dimensions–parts 1-30 to 1-39:Measurement methods and test procedures for mechanical charac-teristics–parts 1-40 to 1-49:Measurement methods and test procedures for transmission andoptical characteristics–parts 1-50 to 1-59:Measurement methods and test procedures for environmental charac-teristics.

60793-1-33 © IEC:2001– 11 –OPTICAL FIBRES –Part 1-33: Measurement methods and test procedures –Stress corrosion susceptibility1 Scope and objectThis part of IEC 60793 contains descriptions of the five main test methods concerning thedetermination of stress corrosion susceptibility parameters.The object of this standard is to establish uniform requirements for the mechanicalcharacteristic stress corrosion susceptibility. Dynamic fatigue and static fatigue tests are usedin practice to determine stress corrosion susceptibility parameters, dynamic n-value and staticn-value.Any fibre mechanical test should determine fracture stress and fatigue properties underconditions that model the practical application as close as possible. Some appropriate testmethods are available:–A:
Dynamic n value by axial tension (see annex A);–B:
Dynamic n value by two-point bending (see annex B);–C:
Static n value by axial tension (see annex C);–D:
Static n value by two-point bending (see annex D);–E:
Static n value by uniform bending (see annex E).These methods are appropriate for types A1, A2 and A3 multimode and type B1 single-modefibres.Static and dynamic fatigue test methods show comparable results if both tests are performedin the same effective measuring time. For dynamic fatigue tests this means a measuring timewhich is (n + 1) times larger than the measuring time of static fatigue tests.When using static fatigue test methods, it has been observed that for longer measuring timesand consequently lower applied stress levels, the n-value increases. The range of measuringtimes of the static fatigue tests, given in this standard, approaches the practical situationbetter than that of the dynamic fatigue tests, which in general are performed in relatively shorttime-frames.These tests provide values of the stress corrosion parameter, n, that can be used forreliability calculations according to IEC 62048.2 Normative referencesThe following referenced documents are indispensable for the application of this document.For dated references, only the edition cited applies. For undated references, the latest editionof the referenced document (including any amendments) applies.IEC 62048, The law theory of optical fibre reliability 1_________1
To be published.
60793-1-33 © IEC:2001– 13 –3 ApparatusSee annexes A, B, C, D, and E for each of the layout drawings and other equipmentrequirements for each of the methods respectively.4 Sampling and specimensThese measurements are statistical in nature. A number of specimens or samples from acommon population are tested, each under several conditions.Failure stress or time statistics for various sampling groups are used to calculate the stresscorrosion susceptibility parameters.4.1 Specimen lengthSpecimen length is contingent on the test procedure used. See the respective annexes A, B,C, D and E for the length required for the test method. For tensile tests, the length rangesfrom 0,5 m to at most 5 m. For two-point bending tests, the actual length tested is less than1 cm and for uniform bending tests about 1 m.4.2 Specimen preparation and conditioningAll of the test methods shall be performed under constant environmental conditions. Unlessotherwise specified in the detail specification, the nominal temperature shall be in the range of20 °C to 23 °C with a tolerance of ±2 °C for the duration of the test. Unless otherwisespecified in the detail specification, the nominal relative humidity (RH) shall be in the range of40 % to 60 % with a tolerance of ±5 % for the duration of the test.Unless otherwise specified, all specimens shall be pre-conditioned in the test environment fora minimum period of 12 h.The use of stress corrosion susceptibility (and proof stress) parameters for reliabilityestimates is still under consideration. A method for extrapolating such parameters to serviceenvironments different from the default environment specified above has not been developed.It has been observed that the value of n produced by these tests can change after even briefexposure of the fibre to elevated temperature and humidity. A guide for the use of thesemethods is documented in IEC 62048.The observed value of stress corrosion susceptibility parameter, n, may differ between fatiguetest methods. Influences on the results have been observed concerning the measuring timeand the applied stress level. Care should be taken in the choice of test method. This shouldbe agreed between the user and manufacturer.5 Reference test methodMethod A is the reference test method and shall be used to resolve disputes because it yieldsminimal values compared to the others and may be completed in a duration practical fordispute resolution.

60793-1-33 © IEC:2001– 15 –6 ProcedureSee annexes A, B, C, D and E, respectively, for the individual test methods.Each of several samples (consisting of a number of specimens) is exposed to one of anumber of stress conditions. For static fatigue tests, a constant stress is applied from sampleto sample and time to failure is measured. For dynamic fatigue tests, the stress rate is variedfrom sample to sample and the failure stress is measured.The following is an overview of the procedures common to all methods:–complete pre-conditioning;–divide the specimens into sample groups;–apply the specified stress conditions to each sample group;–measure time or stress at failure;–complete calculations7 CalculationsThe calculations for each individual test method are found respectively in annexes A, B, C, Dand E.8 Results8.1 The following information shall be reported with each test:–fibre identification;–test date;–stress corrosion susceptibility parameter;–test method.8.2 The following information shall be provided upon request:–specific information as required by the test method;–any special pre-conditioning.Clauses A.5, B.5, C.5, D.5, and E.5 have results that apply respectively for each specificmethod.9 Specification informationThe detail specification shall specify the following information:–information to be reported;–any deviations to the procedure that apply;–failure or acceptance criteria.

60793-1-33 © IEC:2001– 17 –Annex A (normative)Dynamic n value by axial tensionThis method is designed for determining the dynamic stress corrosion susceptibility parameter(dynamic n value, nd) of optical fibre at specified constant strain rates.This method is intended only to be used for use with those optical fibres of which the medianfracture stress is greater than 3 GPa at the highest specified strain rate. For fibres withmedian fracture stress less than 3 GPa, the conditions herein have not demonstratedsufficient precision.This method is intended to test fatigue behaviour of fibres by varying the strain rate. The testis applicable to fibres and strain rates for which the logarithm of fracture stress versus thelogarithm of strain rate behaviour is linear.A.1 ApparatusThis clause describes the fundamental requirements of the equipment used for dynamicfracture stress testing. There are several configurations that meet these requirements.Examples are presented in figures A.1 to A.3. Unless otherwise specified in the detailspecification, use a gauge length of 500 mm for tensile test specimens.Speed-controldeviceMotorVariable
speeddriveCapstan diameter(50 mm min.)To load cellTo cross headFibreFibre
holders(capstans)Load cellGauge length(500 mm min.)IEC
1385/01Figure A.1 – Schematic of translation test apparatus

60793-1-33 © IEC:2001– 19 –FibreNon-rotating capstanRotating capstanwith torsion sensorIEC
1386/01Figure A.2 – Schematic of rotational test apparatusLoad cellVertical non-rotatingcapstanRotating capstanFibreIEC
1387/01Figure A.3 – Schematic of rotational test apparatus with load cellA.1.1 Support of the specimenGrip the fibre length to be tested at both ends and subject the fibre to tension until fractureoccurs in the gauge length section of the fibre. Minimize the fibre fracture at the grip byproviding a surface friction that prevents excessive slippage.Do not include breaks that occur at the grip in the sample or use them in the calculations.Use a capstan, optionally covered with an elastomeric sheath, to grip the fibre. Wrap asection of the fibre that will not be tested around the capstan several times and secure it atthe end with, for example, an elastic band or masking tape. Wrap the fibre with no crossovers.The gauge length is the length of fibre between the axes of the gripping capstans before it isstretched.

60793-1-33 © IEC:2001– 21 –Use a capstan and pulley diameter so that the fibre is not subjected to a bending stress thatcauses the fibre to break on the capstan. For typical silica based fibres, the bending stressesshall not exceed 175 MPa when the fibre is wrapped as shown in the figures or traverses apulley. (For 125/250 µm – cladding/coating – silica fibre, the minimum capstan diameter isthen 50 mm.) Provide a capstan surface tough enough that the fibre does not cut into it whenfully loaded. This condition can be determined by pre-testing.A.1.2 Stressing applicationElongate the fibre at a fixed strain rate until it breaks. The rate of elongation is expressed aspercentage per minute, relative to the gauge length. Two examples for doing this are asfollows:a)increase the separation between the gripping capstans by moving one or both of thecapstans at a fixed rate of speed, with the starting separation equal to the gauge length(figure A.1); orb)rotate one or both of the gripping capstans, to take up the fibre under test (see figures A.2and A.3).The strain rate is the change in length between the two locations, in per cent, divided by thetime.If method b) is used, ensure that the fibre on the capstan does not cross over itself as it iswrapped.If fibres are tested simultaneously, protect each fibre from adjacent fibres so that whiplash atfracture does not damage other fibres under test.A.1.3 Fracture force measurementMeasure the tensile stress during the test and at fracture for each test fibre by a load cell,calibrated to within 0,5 % (0,005) of the fracture or maximum load, for each range of fracturestress. Calibrate the load cell while oriented in the same manner as when testing the fibreunder load. For method b), use a light, low-friction pulley (or pulleys) in place of the non-rotating capstan (see figure A.2), or the rotating capstan (see figure A.3), when calibratingload cells with a string and calibration weight.Use a string, attached at one end to the load-measuring device (or its capstan), to duplicatethe direction of an actual test fibre and be of a thickness or diameter comparable to that of atest fibre. A minimum of three calibration weights are recommended for load cell calibrationwhich bracket the typical fracture or maximum load (50 % below maximum, maximum and50 % above maximum).Recording the maximum tensile load at the time of fracture may be obtained for example by astrip chart recorder. The response time shall be sufficient to report the fracture stress within1 % of the actual value.NOTE
Frictional effects from the pulleys can lead to substantial errors in the load cell calibration of rotatingcapstan testers for horizontally mounted fibre.A.1.4 Strain rate controlDetermine the setting for the speed control unit by trial in order to meet the specified strainrates. Express the strain rate as a percentage of gauge length per unit time. Unless otherwisespecified in the detail specification, the maximum strain rate shall be
equal to or less than100 %/min. Select the actual maximum strain rate by taking into account aspects of the test

60793-1-33 © IEC:2001– 23 –method such as equipment considerations, material properties of the samples, etc. In additionto the maximum rate, use three additional strain rates, each reduced sequentially by roughly apower of 10 from the maximum.It is possible to minimize test duration by using a faster strain rate in conjunction with areduced load. For example, if a strain rate of 0,025 %/min is specified, test some specimensat the next fastest rate (0,25 %/min) to establish a range of fracture stress. Then pre-load to alevel equal to or less than 80 % of the lowest fracture stress found for the initial trialspecimens at the next fastest rate.A.1.5 Stress rate characterizationThe stress rate may vary with fibre type, equipment, breaking stress, fibre slippage, and strainrate. Characterize the stress rate, aσ&, at each strain rate used in the fatigue calculationaccording to:)8,0()(2,0fffaσσσσ×−×=tt&(A.1)whereσfis the fracture stress;t(σf)is the time to fracture;t(0,8 × σf)is the time at 80 % of the fracture stress.A.2 Test sampleA.2.1 Sample sizeBecause of the variability of test results, test a minimum of 15 specimens for each strain rate,and drop the lowest breaking fracture stress data point for each strain rate. Alternatively, ifthe standard error of estimate of slope σf vs. aσ& is 0,0017 or greater (as explained in F.2),test a minimum of 30 specimens for each strain rate and drop the lowest two breaking fracturestress data points for each strain rate.A.2.2 Sample size (optional)As explained in clause A.2.1, additional specimens may be required for some applications inwhich the confidence interval on the estimate of the dynamic (tension) stress corrosionsusceptibility parameter, nd needs to be known. Refer to table F.1 for various sample sizes,depending upon the expected dynamic Weibull slope, md. Appropriate use of the algorithm inclause F2 is restricted to tests in which the same sample size is specified for each strain rate.A.3 ProcedureThis procedure describes how to obtain fibre fracture stress on a given sample set tested at agiven strain rate. Calculations of population statistics are presented in clause F.2.A.3.1Set and record the gauge length (see A.1.2).

60793-1-33 © IEC:2001– 25 –A.3.2Set and record the strain rate (see A.1.4).A.3.3If method a) of A.1.4 is used, return the gripping capstans to the gauge lengthseparation.A.3.4Load the test specimen in the grips, one end at a time. The tangent point of the fibreshall be in the same location as that for the load calibrations. Guide each specimen so thatthe fibre makes at least the required number of turns around the capstan without crossingover itself.A.3.5Re-set the load recording instrument.A.3.6Start the motor to stress the fibre. Record the stress vs. time until the fibre breaks.Stop the motor.A.3.7Repeat steps A.3.3 through A.3.6 for all fibres in the sample set.A.3.8Calculate the fibre fracture stress, σf, for each break. Use equation (A.2).A.3.9Calculate the stress rate, aσ&.A.3.10Complete the required population statistic calculations. Use equations (A.3) to (A.6).A.4 CalculationsA.4.1 Fracture stressThe following method can be used to calculate the fracture stress, σf, when the coatingcontribution is negligible (less than 5 %), such as on common 125 µm diameter fibre with acoated diameter of 250 µm (polymer coating):σf = T/Ag(A.2)whereTis the force (tension) experienced by the composite specimen at fracture;Agis the nominal cross-sectional area of the glass fibre.A more complete method is given in clause F.3 for use when the coating contribution isimportant.A.4.2 Fracture stress at a given strain rateThe following steps are required to form a Weibull plot characterizing the population.a)Sort the fracture stresses from minimum to maximum. Assign a rank, k, to each. Rank isthe order, e.g. first is the weakest, second is the next weakest, etc. Assign a different rankto each break, even if several breaks have the same fracture stress.b)Calculate the cumulative probability of failure, Fk, for each break:Fk = (k – 0,5)/N,
k = 1, 2, . N (A.3)where N is the sample size.

60793-1-33 © IEC:2001– 27 –c)Graph ln [–ln (1 – Fk)] vs ln (σf) to form the Weibull plot.NOTE
Special Weibull graph paper is available for this.d)Label the plot with the required information.For a given gauge length and diameter, the dynamic fatigue Weibull plot is associated withthe following cumulative probability function:Fk = 1 – exp [–(σf/σo)md](A.4)Let k(P) = P × N + 0,5 define a rank associated with a given probability, P.If k(P) is an integer, let σf (P) = σfk (P), the fracture stress of the k(P)th rank. If k(P) is notan integer, let k1 be the integer below k(P) and k2 = k1 + 1.Then, let σf (P) = (σfk1 × σfk2)1/2.The median fracture stress is σf (0,5). The Weibull slope is:()[]()[]15,0 ln85,0 ln46,2ffdσσ−=m(A.5)The Weibull parameter is:()[]+=5,0ln3665,0expfdoσσm(A.6)Graph the Weibull plot for each stress rate, and determine the median fracture stress σf (0,5)for each stress rate.A.4.3 Dynamic (tension) stress corrosion susceptibility parameter, ndThe median fracture stress σf (0,5) as defined in A.4.2, will generally vary with constant stressrate, as follows:daf1loglogn+σσ& + intercept(A.7)where intercept is the log of fracture stress at a stress rate of unity as shown in figure A.4.Intercept can be calculated from the following:intercept = Y – (slope) × X(A.8)Unless otherwise specified, use the algorithm in clause F.2 to calculate X, Y, the estimateof nd, and the 95 % confidence interval for the test. Unless otherwise specified, the standarderror of estimate of slope log σf vs. aσ& shall be less than 0,0017. Refer to clause F.2 todetermine the standard error of estimate of slope.

60793-1-33 © IEC:2001– 29 –A.5 ResultsThe following data shall be provided upon request:–strain rates;–sample size per strain rate;–standard error of estimate;–Xand Y;–gauge length;–test environment;–environmental pre-conditioning time;–fracture stress calculation method;–Young's modulus of fibre (if taken into account);–Young's modulus of coating(s) (if taken into account);–Weibull plots for all strain rates (if used);–method of calculating the stress rate.••••••••••••••••••••••••••••••••••••Log (stress rate)
MPa/minLog (fracture stress)
MPaIntersection••••••••••••Slope =11 + ndIEC
1388/01Figure A.4 – Representation of dynamic fatigue graph

60793-1-33 © IEC:2001– 31 –Annex B (normative)Dynamic n value by two-point bendingThis procedure provides a method for measuring the dynamic fatigue parameters (dynamic nvalue, nd) of optical fibre in two-point bending at a constant platen velocity. This method isintended to test fatigue behaviour of fibres by varying the platen velocity. The test isapplicable to fibres and platen velocities for which the logarithm of fracture stress versus thelogarithm of platen velocity behaviour is linear.B.1 ApparatusA possible test apparatus is schematically shown in figure B.1. This equipment is designed tomeasure the strain/stress required to break an optical fibre in a two-point bending geometryby measuring platen separation at fracture. This technique is readily amenable to various testenvironments.B.1.1 Stepper motor controlThis device allows accurate, reliable, repeatable motorized control of the linear table. Amaximum step length of 1 µm shall be used. A step length of 0,1 µm could be used for higheraccuracy.B.1.2 Stepper-motor-driven moving platenThe moving platen converts the stepper motor rotation to linear translation by means of a leadscrew.B.1.3 Stationary platenThis device holds the fibre against the moving platen.B.1.4 Platen velocityPlace the fibre between two platens that are brought together by a computer controlledstepper motor at a specified constant platen velocity (V = constant) until the fibre breaks.Unless otherwise specified in the detail specification, use velocities 1 µm/s, 10 µm/s,100 µm/s, 1 000 µm/s, each accurate to ±10 %.B.1.5 Fibre fracture detecting systemOne of the following techniques may be used to detect fibre fracture.B.1.5.1 Method 1Use an acoustic emission detector or transducer and computer to sense the fibre break andplaten position at time of break. The computer then stops the platen and displays the platenseparation at the time of the break.

60793-1-33 © IEC:2001– 33 –B.1.5.2 Method 2Incorporate a force (pressure) transducer into the stationary platen and connect it to asuitable signal conditioning equipment to measure force exerted on the fibre during the test.When the fibre breaks the force drops to zero, providing a means of detecting the break.B.1.5.3 Method 3Launching light through a fibre during the test and monitoring the output signal is anothertechnique for detecting fibre fracture. When the fibre breaks, the transmission is lost.With all of the techniques above calculate the platen separation at fracture d as:d = platen starting position – platen travel(B.1)B.2 Test sampleThe test sample is a length of coated optical fibre approximately 30 mm to 120 mm long. Theglass diameter shall be known to ±1 µm and coating diameter shall be known to ±5 µm.Unless otherwise specified in the detail specification, the sample size for each velocity shallbe at least 15 specimens.B.3 ProcedureB.3.1The following is one example of a calibration procedure. Set the distance between theplaten to zero when the faces of the platen are completely touching. When contact is made,the readout on the stepper motor controller should be zero. The platen separation value dwhen the fibre breaks may be verified by checking the distance with a gauge block. The zeroposition should be repeatable to ±5 µm.NOTE
The surfaces of the platen should be carefully cleaned before they are run together for touching.B.3.2Unless otherwise specified in the detail specification, set the initial fibre platenopening gap to 12,00 mm including groove depths.B.3.3Before a population of fibres for a given platen velocity is tested, break an identicalfibre from the same group to determine the platen separation at fibre fracture. This platenseparation d is used to calculate the breaking stress (equation (B.2), (B.3) and B.4)). An initial(starting) platen separation can be determined from equations (B.2), (B.3), (B.4) and (B.5)using a value of stress equal to 50 % of the breaking stress. This will allow the duration of thetest to be reduced and the highest platen velocities to be achieved, since the maximumstepper motor speed may limit the maximum obtainable platen velocities.It is possible to minimize test duration by using a faster platen velocity in conjunction with areduced load. For example, if a platen velocity of 1 µm/s is specified, test some specimens atthe next fastest rate (10 µm/s) to establish a range of fracture stresses. Then preload to alevel equal to or less than 80 % of the lowest fracture stress found for the initial trialspecimens at the next fastest rate.

60793-1-33 © IEC:2001– 35 –B.3.4Carefully grasp both ends of the test specimen, bend it carefully, and insert it betweenthe platen, then pull it upwards to position it as shown in figure B.2. Do not touch the bentfibre (gauge length) with fingers when handling or loading fibres. The apex of the fibre shouldalways be at the same position in the fixture. This minimizes the effect of a non-parallelplaten. Fibre orientation, whether up or down, does not matter.B.3.5After the specimen has broken, brake the stepper-motor to a stop and record theplaten separation at the break.B.3.6Repeat steps B.3.1 to B.3.5 for each fibre sample at the specified load rate, and for allsamples at the other specified load rates.B.3.7Calculate the fibre fracture stress, σf, for each break, using equations (B.2) to (B.4).B.3.8Complete the required population statistic calculations, using equations (B.5) to (B.6).B.4 CalculationsB.4.1 Fracture stressCalculate the fracture stress of each fibre by:()ffof5,01εαεσ×′×+×=E(B.2)gcff2198,1dddd+−=ε(B.3)25,075,0−×=′αα(B.4)whereσf is the fracture stress in GPa;Eo is the Young's modulus (72 GPa);εf is the fracture strain at the apex of the fibre;α is the correction parameter for non-linear stress/strain behaviour (typical value for α is 6);df is the glass fibre diameter in µm;d is the distance between platen at fibre fracture in µm;dc is the overall fibre diameter including any coating in µm;2dgis the total depth of both grooves in µm (see figure B.2).B.4.2 Dynamic (two-point bending) stress corrosion susceptibility parameter, ndThe median fracture stress, σf (0,5), will generally vary with constant platen velocity, V,according to:onintersectilog11(0,5)Logdf+×−=rVnσ (B.5)

60793-1-33 © IEC:2001– 37 –whereris the radius of glass fibre;interceptis the logarithm of fracture stress at a constant platen velocity of unity as shown infigure B.3.Intercept can be calculated from:intersection = Y – (slope) × X(B.6)Unless otherwise specified, use the algorithm in F.2 to calculate X, Y, the estimate of nd,and the 95 % confidence interval for the test. Unless otherwise specified, the standard errorof estimate of slope log σf vs. log V shall be less than 0,0017. Refer to F.2 to determine thestandard error of estimate.B.5 ResultsThe following data shall be provided upon request:–platen velocities;–sample size for each platen velocity;–the standard error of estimate;–test environment;–environmental pre-conditioning time;–Young's Modulus of fibre glass (if assumed other than what is given in F.3);–Weibull plots for all platen velocities (if used);–X and Y;–fibre (glass) diameter.ComputerSteppermotorSteppermotorcontrolFailuredetectionsystemDetectorMovableplatenFibreFixedplatenIEC
1389/01Figure B.1 – Schematic of two-point bending unit

60793-1-33 © IEC:2001– 39 –dGlassCoatingPlatendgdcdgdfIEC
1390/01IEC
1391/01(a)(b)Figure B.2 – Schematic of surface platen••••••••••••••••••••••••••••••••InterceptLog (fracture stress)
MPaSlope =nd – 11Log V/rIEC
1392/01Figure B.3 – Dynamic fatigue data schematic

60793-1-33 © IEC:2001– 41 –Annex C (normative)Static n value by axial tensionThis method is designed for determining the static fatigue parameters (dynamic n value, ns) ofindividual optical fibre lengths under tension. This method is intended to test static fatiguebehaviour of fibres by varying the applied stress levels.C.1 ApparatusPossible arrangements of test equipment are schematically shown in figure B.2. Eacharrangement consists of a means of applying stress to a fibre and monitoring time to fracture.Unless otherwise specified in the detail specification, the gauge length, i.e. the distancebetween the capstans, shall be 500 mm.C.1.1 Gripping the fibre at both endsSee A.1.1.C.1.2 Stressing the fibreThe stress is applied on the fibre by hanging a known weight on one capstan (see figure C.1).Several specimens are tested at a given nominal stress level. The range of actual stresslevels for a given nominal level can influence the quality of the measurement. For the simplemedian computation method, the range of stress levels for a given nominal shall be within±0,5 % of the nominal. For the homologous method and the maximum likelihood estimatemethod, the individual stress levels for each specimen shall be recorded for use in thecomputation. See C.4.2.C.1.3 Measuring time to fractureThere are many techniques to monitor time to fracture which can meet the requirements ofthis test method. One way to monitor the time to fracture is to set up timers underneath thehanging weights used to apply the stress on the fibre.C.2 Test sampleC.2.1 Sample size for each nominal stress levelUnless otherwise specified in the detail specifications, use the sample size for each nominalstress level of at least 15.C.3 ProcedureTest a minimum of five different nominal applied stress levels, σa. Choose the nominalstresses such that the median times to fracture range from about 1 h to about 30 days inroughly equal distance on the logarithmic scale. The loads necessary to achieve this forstandard silica fibres are in the range of 30 N to 50 N.

60793-1-33 © IEC:2001– 43 –Since the time to fracture is dependent on both the fracture stress of the fibre and the fatigueparameter, the actual nominal stress levels applied and the number that are applied can bedetermined iteratively. Alternatively, a broad range of levels may be applied at the beginningof a measurement. Data from test sets that break too soon or take too long to break may bediscarded.Upon completion of pre-conditioning, load the fibres into the unit. Monitor and record the timeto fracture for each fibre fracture. When testing a sample set for a given nominal stress level,as soon as the median specimen has broken the test may be terminated early. That is, if morethan half of the samples have broken, the computation can be carried out and a median timeto fracture determined before all the remaining samples fail. The standard error of theestimate shall be computed and reported for each measurement. Unless otherwise specifiedin the detail specification, the standard error of the estimate shall be less than 1.C.4 CalculationsC.4.1 Fracture stressSee A.4.1.C.4.2 Static (tension) stress corrosion susceptibility parameter, nsUnless otherwise specified, the following method shall be used to determine ns. Alternatively,other methods, for example homologous or maximum likelihood estimate, can be used todetermine ns (see A.4).C.4.3 Simple medianThis method does not require an assumption of linearity of the Weibull slope. Since all thedata are not used, it can produce a larger standard error of the estimate than others. For eachnominal stress level, σi, the median time to fracture, ti, is determined. Fit the data to thefollowing linear regression model by minimizing the sum of squared errors:()()iilninterceptlnstn=+−σ(C.1)The standard error of the estimate for ns is reported by most statistical packages. The medianof ln(σi) and the median of the ln(ti) are also reported. The value of intercept in the aboveequation is as follows:intercept = median [ln(ti)] + ns median [ln(σi)](C.2)C.5 ResultsThe following data shall be reported:–fibre identification;–test date;–static (tension) stress corrosion susceptibility parameter, ns (other parameters are underconsideration).The following data shall be provided upon request:– fibre diameter;– coating diameter (if it is taken into account);– test environment;– gauge length;

60793-1-33 © IEC:2001– 45 –– initial sample size for each nominal stress level and the number of nominal stress levels;– environmental pre-conditioning time, where applicable;– fracture stress calculation method. If the
...