Measurements of lightwave componentreflections with the Agilent 8504B precision reflectometer

Product Note 8504-1

2

The precision reflectometer

A new development in opticalreflectometry, the Agilent 8504Bprecision reflectometer, signifi-cantly extends measurement capa-bilities. The two-event resolutionis better than 25 micrometers,while the dynamic range exceeds80 dB. A measurement tool nowexists specifically for thedesigner and manufacturer ofprecision lightwave componentsand connectors. The sensitivityexceeds that of the best powermeter solutions, while the resolu-tion is sufficient to isolate eachindividual reflection within asmall, complex optical assembly.

Table of Contents

Agilent 8504B operation summary 3

Return loss concepts and measurements 4Basic concepts of reflection 4Problems that result from reflected light 4Survey of return loss measurement methods: 5

Power meters 5Optical time-domain reflectometers 6Optical frequency-domain reflectometers 6

The precision reflectometer technique 6

The general measurement process 8Instrument warm-up 8Reference extension cable selection 8Cleaning connectors 8Select operating wavelength 8Measurement calibration: 9

Balance receiver 9Magnitude calibration 9

Measure the test device 10Optimize the instrument setup 10Measurement example: Connector pair 10

Measurement procedure 10Increasing the measurement rate 11

Measurement example: Characterizing a photodiode assembly 11Measurement procedure 11Optimizing the instrument setup 11

Applications 13Measuring the reflections from an optical isolator 13Characterizing reflections within a laser assembly 13Characterizing devices pigtailed with multimode fiber 14Precision measurements of differential length. A 1XN coupler 15Characterizing high return loss terminations: Index matching gel 16

Common questions and answers 17

Understanding measurement accuracy 18

Bibliography 19

3

Agilent 8504B operation summary

The following one page operatingsummary is intended as a brief ref-erence. Detailed information is dis-cussed within the product note.

1. Warm-up the instrument: Formaximum dynamic range and mea-surement stability, allow the instru-ment to warm up for two hours.

2. Clean all connections: Clean allfiber connectors and instrument testports used in both the calibrationand measurement process.

At this point you can follow the oper-ating procedure below, or use theinstrument’s “Guided Setup” feature.Press SYSTEM, [Guided Setup].

3. Select wavelength: Press PRESET, MENU, and select theappropriate wavelength (under[SOURCE MENU]).

4. Determine reference extensionlength: Measure the length (L2) offiber cable leading to the deviceunder test (DUT). Attach extensioncable L1 between the referenceextension ports with length equal toor slightly less than L2. If the exten-sion cable length L1 is greater thanL2, the device may not be seen in theinstrument’s measurement range.

5. Perform a calibration:This removes DC offsets and polar-ization sensitivity, and sets a cali-brated reference level. Select CAL,[Guided Cal]: Terminate the instru-ment test port with the high returnloss load (>40 dB) supplied with theinstrument. Adjust the polarizationbalance as directed.

Note: Once the polarization calibra-tion has been performed, the refer-ence extension cable L1 and thepolarization adjustment knobs mustremain stationary. Attach a fiber

cable of length L2 or slightly longer,with known return loss (typically aFresnel reflection) to the instrumenttest port. Measure the standard asdirected. If the system is operatingcorrectly, there should be a singleresponse seen, similar to the displayshown below.

6. Connect the test device: Attachthe device under test (DUT) to theinstrument test port. Let the instru-ment complete a full sweep. Locatethe reflections of interest and reducethe measurement span as much aspossible using the MKR FCTN andspan keys.

7. Increase sensitivity: Increased sen-sitivity can be achieved through aver-aging. Press AVG, [AVERAGING ON].The number of averages is set usingthe [Averaging Factor] function. Thenumber of averaged traces is displayedat the left border of the display.

L1 L2

DUT

27 cm (in fiber)measurement

range

0<L2–L1<25 cm

Agilent 8504B Measurement Setup

Typical Display

0 dB return loss

Reflectiondisplayed indB return loss

50 dBreferencelevel

Instrument noise floor

4

Return loss concepts andmeasurement techniques

The reflection coefficient ‘ρ’ of thisinterface is the ratio of the reflectedelectric field to the incident electricfield and is given by the following:

ρ=n1–n2

n1+n2

where n1 is the index of refractionfor the material the light is propa-gating from and n2 is the index ofrefraction of the material the light istraveling to.

The reflectance ‘R’ is a similar termand is defined as the ratio of thereflected beam intensity to the inci-dent beam intensity and is given by:

R= ρ2

= (n1–n2)2

n1+n2

(In the above cases, it is assumedthat the light is traveling normal to the interface plane.)

An example of this phenomenon isan air-to-glass interface. Air has anindex of refraction of 1, while glasshas an index of approximately 1.5.

The reflectance of this interface isthen 0.04 or 4% while the reflectioncoefficient is –0.2. A negative reflec-tion coefficient is an indication thatthe reflected field has experienced a180-degree phase shift relative tothe incident field.

The reflective properties at aninterface can also be described loga-rithmically in decibels. This parame-ter is called return loss (RL) and isgiven by:

RL= –10 log10 (power reflected)power incident

–10 log10 R

The air-glass interface would have areturn loss of 14 dB.

Notice that the smaller the reflec-tion, the larger the return loss. Thisshows the utility of using return lossto describe very small reflections.

Note: In measuring reflections,intuitively you would expect lowreflections to be displayed lower thanhigh reflections. Consequently, theAgilent 8504B uses a different con-vention in displaying return loss.Return loss is computed as:

10 log (power reflected)power incident

as opposed to:

–10 log (power incident )power reflected

Thus a return loss of 20 dB is dis-played as –20 dB and a return lossof 50 dB is displayed as –50 dB.

The majority of this document dealswith components and devices usedin systems that use optical fiber.Another phenomenon that must beconsidered is the coupling of thereflected light back into the fiber. Ingeneral, the light from a reflectiveinterface is not collimated and onlya portion will return back throughthe fiber. Thus, the definition ofreturn loss must be modified again.Return loss is then defined in termsof what actually propagates backthrough the fiber rather than whatwould be predicted solely from thedifferences in the refractive indicesat a boundary.

Problems that result from reflected lightThe most obvious problem thatoccurs when light is reflected is thatthe transmitted signal is reduced. Inthe previous glass-to-air interfaceexample, since 4% of the light wasreflected, only 96% of the originalsignal was transmitted. This corre-sponds to a (–10 log10(0.96)) 0.2 dBloss in power.

Sometimes more important than power loss is the effect thatreflected light has on the perfor-mance of lightwave components.Today’s high-speed lightwave com-munication systems typically usenarrow linewidth lasers. The relativeintensity noise (RIN) and modula-tion characteristics of such lasers canbe significantly degraded by verysmall amounts of backreflected light.Reflected light can also cause “biterrors” in digital communication sys-tems and distortion in analog com-munication systems.

As light reflects off one interface,the reverse traveling waveform maybe re-reflected off another interface(closer to the source). This resultsin two forward traveling waves. Themagnitude of the composite forwardtraveling signal is the vector sum ofthese two waveforms. Depending onthe phase relationship of the twosignals, which is in turn dependenton wavelength and the path lengththat the reflected wave traveled, thetwo waves may add either construc-tively or destructively. Subtlechanges in source wavelength andenvironmental changes (such astemperature) can cause this phaserelationship to vary significantly.Thus, the total power seen at thedetector will vary with time.

For these above mentioned reasons,the components used in lightwavesystems such as isolators, connectors,and photodiodes typically have veryhigh return loss (low reflections)and as few reflections as possible.

Basic concepts of reflectionWhen light travels across the bound-ary between materials with differ-ent indices of refraction or densi-ties, some portion of the light willbe reflected. The figure below showsthe simplest form of this concept, a plane wave traveling perpendicu-lar to the boundary between the two materials.

Boundary

N1 N2

Incident

Reflected

Transmitted

5

Survey of return loss measurement techniques

Contemporary methods

Virtually all return loss measure-ment techniques employ some typeof optical reflectometry. This con-sists of illuminating the deviceunder test and measuring the lightthat is reflected back. Optical reflec-tometers include power meter/cou-pler based systems (sometimescalled optical continuous wavereflectometers or OCWR), opticaltime-domain reflectometers(OTDR), optical frequency-domainreflectometers (OFDR), and theAgilent 8504B precision reflectome-ter. Each of the above mentionedmeasurement techniques offerunique advantages and benefits.

Power meter/coupler based measurements

Return loss measurements can bemade using a power meter such asthe Agilent 8153A lightwave multi-meter and the Agilent 81534A returnloss module. The return loss modulecontains a sensitive detector and a directional coupler. The sourcemodule of the Agilent 8153A illumi-nates the test device, while the direc-tional coupler and detector senseonly the power that is traveling in thereverse direction.

This system is best suited for accu-rate measurement of devices with asingle reflection such as connectors,splices and attenuators. It is alsoused to measure the aggregate or“total” return loss of a device withmultiple reflections.

Return loss when there are multiple reflections

While the power meter system isboth economical and easy to use, itdoes not provide any spatial informa-tion for resolving multiple reflections.There are two important implica-tions to consider. First, identifyingand quantifying each reflection isimportant in the design and manu-facturing of high return loss compo-nents. Second, the total return lossof a component can vary significantlywhen two reflections create a Fabry-Perot resonator.

When multiple reflections exist, thespectral characteristics of the lightsource must be considered. If thelight source has a very narrow line-width, it will then have a very longcoherence length. In brief, thisimplies that the phase characteristicsof the light are very stable. Whenlight reflects off two discrete inter-faces there will be two reverse trav-eling waves. The total reverse travel-ing power will be related to the vectorsum of the two waves. The phaserelationship of these two signals isdependent upon the source wave-length and path length between thetwo reflecting interfaces. The rela-tive phase between the two waveswill then determine whether theywill add constructively or destruc-tively. This is related to the Fabry-Perot effect. An example of the phe-

nomenon can be demonstrated witha simple connector pair with an airgap of 1 mm.

Forward-traveling light will firstencounter the glass-air interface at the end of the first connector.Approximately 3.5% of the light willbe reflected. The majority of the lightwill continue to the air-glass inter-face of the second connector. Again,there will be a 3.5% reflection.

At several discrete wavelengths(such as 1300 nm), the air “cavity”length is such that the two reflectedwaves will be precisely in phase,and the reflected power will be at amaximum. However, at other wave-lengths (such as 1300.4 nm), the twowaveforms will be out of phase, andthe two signals will add destruc-tively. The reflected power will be at a minimum1.

Return loss for two equal reflections1 mm air gap

Ret

urn

Loss

(dB)

Distance (mm)

1 mm

Return loss for two equal reflections1 mm air gap

N1N2

N1L

Er2Er1

ErTotalEr1

Er2 Er2

θ=?Er1

1 In theory, if there were no coupling losses, thestimulus was monochromatic, and all of the sig-nals re-reflected in the cavity were consideredin addition to the primary reflections, thereturn loss can go to infinity, implying a reso-nant condition.

6

The Agilent 8504B precision reflectometerA new development in opticalreflectometry, the Agilent 8504Bprecision reflectometer, signifi-cantly extends measurement capa-bilities. The two-event resolution isbetter than 25 micrometers, whilethe dynamic range exceeds 80 dB. Ameasurement tool now existsspecifically for the designer andmanufacturer of precision light-wave components and connectors.The sensitivity exceeds that of thebest power meter solutions, whilethe resolution is sufficient to isolateeach individual reflection within asmall, complex optical assembly.

The Agilent 8504B precision reflec-tometer is based on the Michelsoninterferometer and uses the tech-niques of “white light” interferome-try. A 1300 nm or 1550 nm low-coherence light source is sent to apower splitter. One path leads to theDUT, while the other path leads to areference mirror. Light reflected offboth the reference mirror and theDUT is recombined and detected. Ifthe path length from the source tothe reflection in the DUT is the sameas from the source to the mirror, acoherent interference signal appearsat the detector. By moving the posi-tion of the reference mirror, theinstrument can then “scan” the testdevice for reflections over a 400mm range (equivalent air distance).

It is worth mentioning again thatthis effect will only occur when ahighly coherent source is used. Thishas some important implications.

The return loss of a component canvary significantly depending uponthe stimulus. Return loss testing of acomponent may not accurately pre-dict its behavior in a working systemif the source characteristics are notsimilar in both situations. In addi-tion, when components are used in asystem with a high coherence laser(such as a DFB or Nd.YAG), the com-ponent return loss characteristicsmay change dramatically as wave-length and temperature are evenslightly varied.

To more completely understand the reflection characteristics ofcomponents and subassemblies, the individual reflections must beisolated and characterized. Othermeasurement techniques are used toboth locate and quantify individualreflections.

The OTDR

Optical time-domain reflectometryis the most familiar and commonlyused reflectometry measurement forthe installation and maintenance ofboth long- and short-haul fiberlinks. OTDRs locate faults by prob-ing a fiber with an optical pulse trainand measuring the reflected andbackscattered light. OTDR’s are typi-cally not used for component-levelmeasurements due to limitations in resolving small or closely spacedreflections, unless very shortimpulses and very high-speed, sensitive detectors are employed.

The OFDR

The two-event resolution and dead-zone problems inherent with OTDR’scan be improved by using a sweptmodulated lightwave instead ofpulses of light. The amplitude andenvelope phase response is recordedover a wide frequency span. Aninverse Fourier transform is per-formed on this data to yield the time-domain response. Depending uponthe modulation bandwidth of theinstrument, the two-event resolutioncan be better than 10 mm (20 GHzbandwidth). The receiver is syn-chronously tuned to the source,thus decreasing the susceptibility tonoise and increasing the dynamicrange to levels near 40 dB. (Seeproduct specific literature for theAgilent 8702 and Agilent 8703).

Precision lightwave reflectometerblock diagram

Detector Display

DUT

1300

1550

WDM

Coupler

Reference Mirror

ReferenceExtension

Return loss, two equal reflections vs. Wavelength

Ret

urn

Loss

(dB)

Wavelength

∆Lambda0.5 mm

1300

A plot of return loss versus wave-length shows this result graphically.This effect is repetitive as a functionof wavelength so that maximumsand minimums will occur at severalwavelengths.

7

(The 400 mm measurement windowcan be offset by simply addinglength to the reference path at theinstrument front panel).

Note that in this measurementscheme there are no pulses of lighttransmitted as there are for theOTDR, and the light is not modu-lated as it is for the OFDR. It is theshort coherence length of the LEDsource in the precision reflectome-ter which leads to very high resolu-tion measurements.

To demonstrate the utility of such aninstrument, consider a high-speedphotodiode.

The physical dimensions of the device are much less than a cen-timeter, yet there are several inter-faces within the device, each poten-tially generating a reflection andcontributing to the overall returnloss of the device. Very high resolu-tion is required to locate, identifyand quantify each reflection. Theprecision reflectometer easily per-forms this task, as seen in the following measurement.

LensDiode chip

Heat sink

In this measurement, return loss isdisplayed versus the one-way pathdistance. The measurement span is 6 mm so the horizontal scale is 0.6mm per division. Return loss is dis-played in decibels. The top of thedisplay is 0 dB return loss, the cen-ter of the display is –50 dB. Higherreturn loss levels are displayed as a lower value on the display sincethey correspond to lower values ofreflection. (Intuitively, you wouldexpect low reflections to be displayedat the bottom of the screen). Thus a51.9 dB return loss value is dis-played as –51.9 dB as noted by themeasurement marker, which corre-sponds to the reflection at the frontface of the diode chip.

The first reflection is the end of the fiber connector. It approaches a Fresnel reflection since the fiberdoes not physically contact thedevice. The reflections off the frontand back of the lens, and the frontand back of the diode chip are clearlyseen. From this measurement, yousee that the return loss of this deviceis dominated by the front face of thediode chip. However, a power metermeasurement would be dominatedby the reflection off the end of thefiber, making it difficult to extractany information about the photodi-ode itself.

It is also important to note the highdynamic range that the interferome-ter technique offers, several ordersof magnitude beyond the capabilitiesof traditional methods.

The measurement range of the precision reflectometer is determinedby the length that the reference mir-ror can travel and is 400 mm (equiva-lent air distance). If measurementsbeyond the 400 mm window arerequired, the window is simply off-set with the appropriate referenceextension patch cord.

Knowing the magnitude and spacingof the reflections yields informationthat is useful in determining compo-nent performance in systems withnarrow linewidth lasers. Specifically,how much the return loss may varyfor a given change in wavelength.

In summary, there are several tech-niques for measuring return loss.The easiest method is to use a powermeter. When reflections need to bespatially resolved, the OTDR is usedfor coarse measurements of longlines. The OFDR can be configuredfor long-line measurements or close-in measurements, depending on themodulation bandwidth used. Forthe highest resolution and dynamicrange, the Agilent 8504B precisionreflectometer technique is optimum.

General measurement processwith the Agilent 8504B

8

ing measurements. This will ensurethat various DC offsets within theinstrument have stabilized and canbe effectively removed from subse-quent measurements.

Once the instrument has beenwarmed up, the “Guided Setup”feature can be used.

Clean fiber ends and instrument test portsIt is a good measurement practice toclean fiber interfaces before per-forming calibrations and makingmeasurements. Clean fiber endsand connector ports are essentialfor good measurements. All threeinstrument ports and any fiber ends(including the reference extension)should be kept clean. Dirty connec-tors can result in spurious responsesand reduced dynamic range. To cleanthe instrument ports, the connectoradapters are removed from theinstrument front panel, exposing thecable ferrule.

Caution: Extreme care should betaken to avoid damaging theinstrument connector ferrules.Damaged connectors reduce mea-surement integrity and are not userserviceable.

Please refer to the connector caredocument in the manual for connec-tor cleaning and care.

Select operating wavelengthThe standard configuration for theAgilent 8504B includes both 1300 nmand 1550 nm measurement capabil-ity. When not using Guided Setup,to select between 1300 or 1550 nmoperation, press MENU, [SourceMenu], and either [1300] or [1550].

Select reference extensionThe first step in making a measure-ment is to select the appropriatelength of reference extension cable.In general, the reference extensioncable length should be equal to the “pigtail” or path length to thedevice to be tested. The length of thereference extension may be slightlyshorter than the pigtail, but shouldnot be longer. If the reference exten-sion is slightly longer than the DUTcable, some of the events to beexamined may not be in the 400 mm(air) measurement span of theinstrument.

The Agilent 8504B is designed so that if the reference extensionlength is identical to that of theDUT fiber, the first response willappear approximately one divisionto the right of the left edge of thedisplay (in a 400 mm span). Thismeans that under nominal condi-tions, the reference extension cableshould be less than 10 millimetersequivalent air length (7 mm actualfiber length) longer than the DUTfiber. Due to the 400 mm allowablemeasurement span, the extensionshould be no more than 360 millime-ters (250 mm actual fiber length)shorter than the DUT fiber. In mathe-matical terms, this is given by:–7 mm<L2–L1<250 mm where: L2 isthe DUT fiber length and L1 is thereference extension length.

The ideal condition is to have thetwo cables be of equal length.

The Agilent 8504B option 001 con-tains reference extension cables of40, 50, 75, 100, 125, 150, and 175 cmlengths. Optimum reference extensionlength depends upon the pathlength to the DUT.

If the DUT has no pigtail, L1 and L2are any two cables of equal length.

A condensed summary of the mea-surement procedure is found in thefront of this document. In operatingthe instrument there are “hardkeys”and “softkeys”. Hardkeys are thosekeys whose function is printeddirectly on the physical keypad.These keys are noted with bold typesuch as PRESET. Softkeys are thosethat are located to the right of theinstrument display and whose func-tion is displayed on the instrumentdisplay. They are noted in bracketssuch as [MKR ZOOM].

There are seven steps to performinga measurement.

1 Warm up the instrument2 Clean fiber ends and

instrument test ports3 Select operating wavelength4 Select reference extension5 Perform a calibration6 Connect the DUT7 Optimize the measurement

The Agilent 8504B has a “GuidedSetup” feature. Guided Setup is apowerful user interface that leadsusers through the steps required tomake measurements. Guided Setupis implemented by pressing SYSTEM,[Guided Setup]. The instrument willthen display each step required tosetup and calibrate the instrument.

The following text describes thesteps used in Guided Setup, as wellas some discussion on why each stepis performed.

Warm-up the instrumentThe Agilent 8504B is capable of mak-ing measurements of extremelysmall reflections. Even very smallspurious signals within the instru-ment can degrade dynamic range.To ensure maximum dynamic rangeand measurement stability, turn onthe Agilent 8504B and allow it towarm up for two hours prior to mak-

9

Option 001 Cable connection guide

DUT path Ref ext L2 (cm) cables L1

40<L2<65 4050<L2<75 5075<L2<100 7590<L2<115 40+50100<L2<125 100115<L2<140 40+75125<L2<150 125140<L2<165 40+100150<L2<175 150165<L2<190 40+125175<L2<200 175190<L2<215 40+150200<L2<225 50+150215<L2<240 40+175225<L2<250 50+175240<L2<265 40+50+150250<L2<275 75+175265<L2<290 40+50+175275<L2<300 100+175290<L2<315 40+75+175300<L2<325 175+125315<L2<340 40+175+100325<L2<350 175+150340<L2<365 40+175+125350<L2<375 75+100+175365<L2<390 40+150+175375<L2<400 50+150+175

Measurement calibrationOnce the reference extension hasbeen selected and attached to theinstrument, the instrument must gothrough a measurement calibration.The measurement calibration con-sists of balancing the polarization-diversity receiver and calibrating theinstrument for accurate reflectionmagnitude measurements.

Balance receiver

As light travels through single modefiber, its polarization characteristicsvary. The magnitude of the detectorresponse is potentially a function of the polarization of the reflectedwaveform relative to the light in thereference arm. Ideally, the detectorresponse is only a function of thereflection magnitude. To ensure that

the reflection measurement is insen-sitive to polarization transforma-tion, a measurement calibration pro-cess is used. The instrumentreceiver consists of two photodiodeswhich respond to orthogonal statesof polarization. During the BalanceReceiver calibration, the instrumenttest port is terminated with a highreturn loss optical load (greater than40 dB) which is supplied with theinstrument. Therefore the light thathits the two detector diodes is onlyfrom the reference mirror. The polar-ization of this light is adjusted withthe polarization adjustment knobs atthe instrument front panel in such away that the responses from eachdetector are equal or balanced.

Note: Polarization of the light in thereference path must not be alteredonce the Balance Receiver calibra-tion has been performed. The refer-ence extension cable and polariza-tion adjustment knobs must not bemoved to ensure optimum perfor-mance. If the reference extension ismoved, the receiver will no longer bebalanced and subsequent measure-ments may be in error.

To perform the Receiver Balancestep, simply follow the instructionsgiven by the instrument.

Magnitude calibration

The magnitude calibration is a sim-ple process consisting of measuringa known reflection. The instrumentthen automatically scales the mea-sured response so the true value isdisplayed.

For every reference extension cablesupplied in the Agilent 8504B option001, there is a corresponding fiber ofequal length which may be used as acalibration standard. The return lossof the fiber end (a super PC ferrule)is 15 dB or 3.16% reflection at 1300nm and 14.7 dB or 3.37% at 1550 nm.

As you continue with the “GuidedSetup” procedure, the instrumentwill display the instructions to per-form the magnitude calibration. TheAgilent 8504B scans through theentire measurement span and deter-mines the value of the peak response,which should be the reflection gen-erated at the end of the cable. If youuse a reflection standard other thana Fresnel reflection, it must have areturn loss greater than 14 dB toprevent saturation of the receiver.

The Agilent 8504B compares themeasured value to the value enteredby the user. Any differences are dueto systematic errors in the measure-ment system. This error term is sub-sequently removed from all furthermeasurements until another calibra-tion is performed or the instrumentis preset to the default settings.

As a check, the calibrated measure-ment of the reflection standard canbe compared to the known value.

It is important to note that there is no length calibration process andconsequently the Agilent 8504B doesnot make absolute length measure-ments directly. It is a common mis-conception that the measurementcalibration process will offset theposition of the reflection standardto the 0 length position. Recall thatwhen the length of the DUT fiber is identical to that of the referenceextension, the first event appears atabout the 40 mm point and not 0 mm. The instrument has no knowl-edge of the length of the referenceextension cable used, nor the lengthof fiber to the device under test.Therefore, all distance accuracy is interms of relative distance to otherreflections in the measurement span.

10

Connect and measure the test deviceThe calibrated instrument can now beused to measure devices. Once thedevice is connected, it must belocated on the screen. If a device isconnected while the instrument is ina measurement sweep, a responsemay be generated in the process ofmaking the connection that is notrelated to the actual device response.Consequently, it is a good measure-ment practice to restart the mea-surement as soon as the device isattached. Press MENU, [Full span],MEAS, [Meas restart]. The instru-ment will then scan over its fullmeasurement range and display all detectable reflections within a400 mm (air) span.

Optimizing the measurementFor measurements at 1300 nm, thereference mirror in the Agilent8504B travels at a constant velocityof 18 mm/sec (21 mm/sec at 1550nm). A 0 to 400 mm sweep will takeover 22 seconds. The sweep-to-sweeptime can only be reduced by reduc-ing the measurement span.Reducing the measurement span willalso increase the spatial resolutionof multiple reflections. It is recom-mended that once the DUTresponses have been located, toreduce the measurement span to thenarrowest range that includes theevents of interest.

Narrowing of the span can beachieved using the MKR FCTNmenu and [MKR ZOOM] key, or byusing the SPAN, CENTER, STOP,and START keys.

Once the measurement span hasbeen optimized to include all eventsof interest, the noise floor can bereduced through data averaging.Averaging reduces the effects ofrandom noise, and can increase themeasurement range by 6 or 7 dB.(Note: the narrower the measure-ment span, the more effective aver-aging becomes). Press AVG and[Avg on]. The default averaging factor

is 16, meaning that each new mea-surement is weighted by a factor of1/16 and will contribute this value tothe current measurement trace. Theaveraging factor can be set to anyinteger value between 2 and 999.

Measurement example:connector pairIn this measurement example, youwill measure the return loss of asimple connector pair. In addition,you will see the connector character-istics as it makes the transition to afully torqued connection.

Measurement procedure

Using the same general measure-ment process described above, selectthe reference extension cable. Thedevice to be measured is a simpleconnector pair. One of the connec-tors is at the end of a 75 cm patchcord. Since the path length to theDUT is simply the length of thispatch cord, the ideal length for thereference extension is 75 cm. Thereference extension could be as shortas 51 cm, which would place thereflection at the end of the measure-ment range, but should not be anylonger than 75 cm.

To calibrate, follow the Guided Setupor press CAL, [GUIDED CAL],attach the supplied high return losstermination at the test port, adjustthe polarization adjustment knobsfor a balanced display as instructedby the instrument and press [DONE].For the magnitude calibration, thecalibration standard can be one ofthe 75 cm cables supplied with theinstrument, or the DUT patchcordconnector end (if the return loss isknown). Connect the calibrationstandard to the test port. Press[FRESNEL 3.16%] when using thesupplied cable or [USER STD] andenter the value for the reflection.

Press [MEASURE]. The analyzer willthen measure the standard andadjust its measurement to coincidewith the actual reflection value.

The patch cord to be tested can now be connected to the test port.Once it is connected, press MENU,[FULL SPAN]. The instrument willthen bring the reference mirror toits starting position and begin anew measurement trace. The posi-tion of the mirror is indicated by asmall red dot at the bottom of theinstrument display. The location ofthe dot on the instrument display isproportional to the location of themirror on the translation stage. Forinstance if the start position of themeasurement is set at 0 mm and thestop position is set at 100 mm, themirror will then be traveling backand forth over the first 25% of the available mirror movement.The mirror position indicator dotwill then move back and forth fromthe left edge of the screen to a point25% or 2.5 divisions away from theleft screen edge. The actual datameasured will be displayed acrossthe entire screen.

As the mirror travels across its full400 mm range, the reflection from theend of the 75 cm patch cord shouldbe seen at approximately the 40 mmpoint. Because the connector is notterminated, a return loss of about–15 dB should be seen.

11

Increasing the measurement rate

To increase the measurement rate,the measurement span must bereduced. Press MKR FCTN, [MAXSEARCH], [MKR ZOOM]. Themarker zoom function places theevent indicated by the marker (inthis case the maximum response) tothe center of the screen and sequen-tially decreases the measurementspan each time the function is acti-vated. Repeat the process until thespan is 20 mm.

Next, examine the return loss as thesecond half of the connector pair ismated with the first. As the secondferrule approaches the first, you seetwo reflections, spaced 2.5 mmapart. The reflection off the secondferrule is low because not all of thereflected light is coupled into thefiber.

As the two ferrules are broughtcloser together, the second reflec-tion appears larger as the couplingefficiency increases. Notice also thatre-reflections are also seen to theright of the second reflection. Theair gap between the two ferrules isnow only 32 microns.

As the connector pair is fullytorqued, the two ferrules makephysical contact and the two reflec-tions become one. The return lossincreases to approximately 40 dB,which is a function of the type ofconnector tested and its cleanliness.

Although this example is of a verysimple device, it shows the twobasic measurement parameters ofthe precision reflectometer, specifi-cally high spatial resolution andhigh dynamic range. The followingexample shows the usefulness ofthese measurement capabilities.

Measurement Example:Characterizing a photodi-ode assembly In this example you will measure acomponent that is physically verysmall, yet has several optical inter-faces, each generating a reflection.

Measurement Procedure

The procedure for measuring thisdevice is virtually identical to thatused for the connector pair. Becausethe photodiode does not have a pig-tail, we simply connect it to the end of the test port patchcord. Theresponse will be more interestingdue to its more intricate construction.

Follow the connector pair measure-ment example to the point wherethe end of the connector is firstmeasured. At this point connect the photodiode. The display shouldthen show several closely spacedresponses.

Optimizing the instrument setup

To speed up the measurement rate and increase the measurement resolution, the measurement spanshould be decreased to the smallestspan that still displays all theresponses of interest. Narrowing of the span can be achieved using the SPAN, CENTER, STOP, andSTART keys, or by using the keysunder the MKR FCTN menu.

12

In this measurement, there are several reflections, some which arevery small. Through averaging, themeasurement sensitivity can beincreased. Press AVG, [AveragingOn], and let the instrument take sev-eral sweeps. This allows us to seereflections that may not have beenvisible before. In this measurement,a fifth reflection is now clearly visi-ble.

Measurement example: Highreturn loss air-gap connector

An example of this is a very highreturn loss beveled-edge connector.The connector response is easily seen.

For this particular connector, younot only measure the return loss, bydecreasing the measurement spanto 1 mm, you measure the gapbetween the connector ferrules.

Press MKR FCTN, and use the knobto move the marker slightly to the left of the first event on thescreen. Press [Mkr–> Start]. Thenuse the knob to move the marker tothe right of the last event on the dis-play. Press [Mkr–> Stop].

13

Applications

The following sections demonstratehow the Agilent 8504B precisionreflectometer can be used to mea-sure a variety of lightwave compo-nents. Although the measurementexamples are just a small sample of potential measurements, they do provide a diverse survey of theinstruments capabilities and thetypes of measurements that can be made.

In virtually all cases, the processes for performing the measurementsare the same as those described inthe previous sections for measuringthe photodiode assembly and theconnector pair. When measurementtechniques not previously intro-duced are encountered, these will bedescribed in appropriate detail.

Measuring the reflections from an optical isolatorIsolators are used to minimize theimpact of reflected light on othercomponents, in particular narrowlinewidth high-speed lasers. Notonly must the isolator perform thefunction of attenuating backre-flected light, it must do so withoutgenerating reflections of its own.

In general, isolators consist of avariety of components includinglenses, crystals and Faraday ele-ments. Each individual componentmay generate a reflection.

The procedure to measure the isola-tor is similar to that used for thephotodiode in the previous sectionunder “Measurement example:Characterizing a photodiode assem-bly.” Once the instrument has beenconfigured and calibrated, the isola-tor is connected and the displayoptimized.

This device has several interfaces, sothere are several reflections. Themeasurement span is 30 mm. Thelargest reflection, generated fromthe angled fiber end, is approxi-mately 60 dB. Beyond the Faradayelement, there are no reflections,indicating that the isolator is per-forming its intended function.

Characterizing reflections within a laser assemblyAs mentioned earlier, lightwavesource performance can be degradedby backreflected light. Measuringthe reflections by probing back intothe laser can give us insight intohow light reflects internally as wellas how back-reflected light may bere-reflected from the laser.

The device shown below will first be measured according to the previ-ous procedures for measuring components.

This particular laser module con-sists of a protective window, a balllens, and the laser chip. During nor-mal operation of the laser, light prop-agates from the chip and is focusedand aligned through the ball lensbefore leaving through the windowinto an attached fiber. Light travelingback into the laser will follow a simi-lar path.

The resulting measurement shows thereflections generated at each inter-face. The largest reflection is thefiber end. Marker 1 shows the reflec-tion from the front of the window, 2,the back of the window, 3, the frontof the ball lens, and 4, the back ofthe laser chip.

The response between marker 3 and4 indicates where the back of theball lens and front of the laser chipmeet. Zooming in, we will examinethis region.

Lens Laser chip

Window

14

Here we see that there are actuallytwo reflections since there is a small gap between the lens andthe laser chip. The size of the gap is 43 microns.

Part of the transmission path of thisdevice is the semiconductor mate-rial of the laser. This material ismore transparent at 1550 nm than at1300 nm. By making the measure-ment at 1550 nm (this requires a newmeasurement calibration), thereflected energy from the end of the laser chip experiences lessattenuation and is then effectivelylarger.

Typically, the transparency of the semiconductor material willvary with the bias current throughthe laser. However, if the laser isbiased above threshold, the trans-mitted energy may be sufficientenough to saturate the receiver of theAgilent 8504B. Therefore, care mustbe taken in setting the laser currentto optimize the tradeoffs betweenmaterial transparency and instru-ment saturation level.

Characterizing devices pigtailed with multimode fiberThe Agilent 8504B precision reflec-tometer measurement systemdetects reflections when the pathlength to the reflection is identicalto the path length to the referencemirror. Because the light travelingthrough multimode fiber will propa-gate over many different paths, themeasurement of reflected energythrough multimode fiber will be dif-ferent than if the device were pig-tailed with single mode fiber.

In essence, the reflection responsestend to be broadened. An example ofthis would be to measure theFresnel reflection at the end of a1.25m length of 62.5/125 multi-mode fiber compared to a similarmeasurement with 9/125 singlemode fiber.

Two things are apparent in theabove plot. First, the response is sig-nificantly broadened. The 3 dBwidth is about 75 microns, comparedto less than 15 microns for the singlemode case. Second, the amplitude ofthe reflection is reduced. This is dueto two factors. The total energy ofthe reflection is distributed over awider range, which decreases thepeak amplitude. In addition, a signif-icant amount of the reflected energyis lost at the multimode to singlemode core mismatch at the instru-ment test port.

The above measurement is for a spe-cific length of multimode fiber. Asthe length of fiber increases, so doesthe effective pulse spreading. Inaddition, for multimode fiberlengths much beyond one meter, themultimode effect will generate multi-ple responses for a single event.

It is interesting to examine theimpact that this spreading has onmeasurements. Consider the photo-diode measurement introduced onpage 11.

The above plot is a composite measurement of the same device.One measurement is with singlemode fiber, while the other is with a 1.25m length of multimode fiber.Each reflection is still visible, andthe relative magnitude information isstill valid.

Although the measurement capabil-ity of the Agilent 8504B degradeswhen using multimode fiber, thedynamic range and two-event reso-lution provide very useful informa-tion in locating and identifyingreflections.

15

The function of the coupler is todivide the input power into twopaths. In some applications, it isdesirable to have the two pathlengths as closely matched as possi-ble. By monitoring the positions of the reflections at the coupler out-puts, we can determine the differen-tial path length.

The measurement is straightfor-ward. After setup and calibration,the coupler is connected to the sys-tem. The output ports of the couplerare left unterminated, which gener-ates two large reflections.

Determining the differential pathlength is a simple matter of placing a marker on each reflection. PressMKR FCTN, [Max Search],[Mkr–>Fixed Mkr], [MKR 2], MKRFCTN, [Peak search], [Next highestpeak]. The analyzer will then displaythe one-way distance betweenmarker 1 and marker 2. The mea-sured distance value is the equiva-lent distance traveled in air, whichin this case is the differential pathlength to the two output ports. Todetermine the physical path lengthdifference, the index of refractionvalue must be entered into theinstrument. In the case of glassfiber, the index of refraction is 1.46.Press MEAS, [N VALUE], and enter1.46 x1. The displayed distance isnow the true one-way physicallength difference.

In order to identify each path, wesimply terminate one of the ports.As expected, one of the measuredreflections will decrease in magni-tude thus indicating which responsecoincides with the terminated port.

This procedure can be used for virtually any 1xN coupler. The mea-surement limitations are sensitivityand two-event resolution. As thepower is divided into more andmore paths, the magnitude of thereflected energy from each outputport will decrease. However, thesensitivity of the Agilent 8504B issuch that the power could be dividedinto over 1000 ports and the reflec-tions still be detectable.

Thus the Agilent 8504B is sensitiveenough to measure virtually any 1xN coupler. Obviously, the practicalmeasurement limitation thenbecomes the two-event resolution.As the differential length of twopaths get smaller and smaller, thereflected signals displayed by theinstrument will eventually overlap.The two-event resolution is 25 microns at 1300 nm and can be65 microns at 1550 nm. Therefore,differential path lengths as small as25 microns (air distance) can bemeasured.

Precision measurementsof differential length:1xN coupler The Agilent 8504B is capable of mak-ing relative length measurementswith very high resolution. This capa-bility can be used to measure rela-tive differences in path length. Anexample of this would be to measurea simple 1x2 coupler.

Out 1

Out 2

In

16

The differential distance measure-ments can also be interpreted in a“time” mode format. This is used toindicate the differential time of flightor propagation time between events.Time format can be activated bypressing FORMAT, [TIME].

Using the time format, we can nowdetermine the effective delay thatone signal path experiences relativeto others. With 25 micron two-eventresolution, propagation delays assmall as 80 femtoseconds can becharacterized.

Characterizing high returnloss terminations: indexmatching gel Index matching gel is commonlyused to terminate an optical fiber inan attempt to minimize reflected lightgenerated at the fiber end. Once thegel has been applied to the fiberend, the common assumption is thatvirtually all of the light is scatteredor “absorbed” and essentially no lightis reflected back.

The high sensitivity of the Agilent8504B allows us to see the real perfor-mance of matching gel. The resultscan be surprising.

The measurement is quite simple.The Fresnel reflection at the end of a cable is located and the instru-ment span is reduced to 1 mm.Then the end of the fiber is dippedin the matching gel. Typically, theresult is that the single reflection is reduced to a return loss valuebetter than 50 dB.

However, there are cases when tworeflections exist, one at the glass/gelinterface and a second reflection atthe gel/air interface.

In most applications, the gel termi-nation provides an adequate fibertermination. However, it is not safeto assume that it completely elimi-nates backreflections.

17

Common questions and answers

The following section deals with avariety of commonly asked questionsabout the Agilent 8504B precisionreflectometer and the techniquesused to make measurements.

Q. Can the Agilent 8504B detectfiber backscatter?

The coherence length of the Agilent 8504B source is approxi-mately 10 to 15 um. The energyreflected from the DUT must havebeen generated within this length inorder to be detected. Although therewill be backscatter generated overthis length, the amount of reflectedenergy is too small to be detected bythe Agilent 8504B.

Q. Can the Agilent 8504B measurefusion splices?

The measurement sensitivity of Agilent 8504B allows reflectionswith return loss values up to 80 dB tobe measured. In general, this is notsufficient to detect good fusionsplices.

Q. The Agilent 8504B displays 401measurement points in any mea-surement span. Is it possible for anevent to be between these pointsand therefore not be detected?

The Agilent 8504B is designed tomake a measurement every 2.5microns along the measurementpath, independent of the measure-ment span. The coherence length ofthe internal source is 10 to 15microns. Thus several points will bemeasured within this coherencelength. Consequently a reflectionwill always be detected, even if itfalls between the actual points ofdetection.

For wide measurement spans, there are still only 401 data pointsdisplayed, yet measurements aremade every 2.5 microns. For a 10 cmmeasurement span, there will be

40,000 measurements made. A pointwill be displayed every 250 microns(401 points over a 10 cm span).Thus each displayed point will rep-resent the largest response detectedover 100 actual measurements. Thisalso indicates why the highest two-event resolution is achieved in thenarrowest measurement spans.

Q. Can the Agilent 8504B be used tomeasure devices pigtailed withmultimode fiber?

The Agilent 8504B can be used tomake measurements in multimodefiber. However, amplitude accuracyand spatial resolution will bedegraded. See “Characterizingdevices pigtailed with multimodefiber” on page 14.

Q. What happens when devices with polarization maintaining fiber(PMF) are measured?

Depending upon how light islaunched into PMF, it will travel atdifferent velocities. Light travelingon one principal axis will travel at a different rate than light travelingon the other principal axis. The lightemitted from the Agilent 8504Bsource is only partially polarized,thus light will typically be travelingon both axes. Thus for a single reflec-tion, two extra responses mayappear.

Q. Will the moving mechanical com-ponents wear out?

The gears and motors in the Agilent 8504B are all designed towork with much heavier loads andstresses than what is actuallyrequired. However, to avoid unnec-essary wear, the Agilent 8504B willautomatically go into trigger holdmode and stop sweeping wheneverthe instrument makes 500 sweeps(or the specified number of aver-ages, whichever is greater) withoutthe operator pressing a front panelkey. Simply press MEAS,[Continuous], to resume sweeping.

Q. Why are the interferometerfringe patterns not seen?

An envelope detector is used tosmooth out the fringe patterns, soonly one response within the sourcecoherence length is displayed.

Q. What is the difference betweenmeasurements made with a powermeter versus the Agilent 8504B andcan I determine total return lossusing the Agilent 8504B?

Total return loss and spatiallyresolved return loss were discussed in“Survey of return loss measurementtechniques” on page 5. Total returnloss is a function of the source char-acteristics as well as the reflectionmagnitudes and their spacing. The8504B can yield insight into thecomponents of total return loss, butnot give an actual value.

Q. What effects do connectors orother losses have when placed infront of the DUT?

Losses will reduce the measurementsensitivity of the Agilent 8504. Ifthese losses are not included in themagnitude calibration, they willalso decrease the measured value of reflections.

Q. What happens when I measure anarrowband device such as a WDMfilter?

The high spatial resolution of theAgilent 8504B is based upon the widespectral width and subsequentshort coherence length of its LEDsource. If a test device’s responsevaries versus wavelength over thespectrum of the source, amplitudeaccuracy and spatial resolution maybe degraded.

18

UnderstandingMeasurement Accuracy

There are several elements whichaffect the accuracy of any measure-ment made with the Agilent 8504BPrecision Reflectometer. In general,various contributions to measure-ment error include systematic errorsthat are inherent in the PrecisionReflectometer measurement tech-nique and items that degrade mea-surement accuracy, yet can be mini-mized by the user through good mea-surement techniques.

Measurement errors can also beclassified into those that affectamplitude measurement accuracyand those that affect positional orspatial accuracy.

Amplitude accuracy issuesThe uncertainty in the Agilent8504B measurement of return lossmagnitude comes from several dif-ferent factors including:

• Dynamic accuracy• Connector repeatability• Transmission path losses• Measurement flatness vs. position• Sweep-to-sweep repeatability• Polarization sensitivity• Calibration accuracy• Chromatic dispersion effects• Source spectral width

Dynamic accuracy

Dynamic accuracy refers to the abil-ity of the Agilent 8504B to accu-rately measure over a wide range of return loss values. The Agilent8504B is typically calibrated at thelevel of a Fresnel reflection. Ideallythe response of the receiver block inthe Agilent 8504B will be linear versusreturn loss. However, as the reflectedsignals from the DUT become verysmall, the response from the detec-tor can deviate from this ideal linearrelationship. Optimization of thereceiver has minimized this effect.The resulting contribution to ampli-tude measurement accuracy isdependent upon the reflection mag-nitude. It is typically less than ±1.5

dB, but degrades as the reflectionmagnitude approaches 80 dB. Referto the curves in the specificationtables of the Agilent 8504B operatingmanual for detailed information.

For measurements of small reflec-tions, dynamic accuracy can be asignificant uncertainty term.

Connector repeatability

There will always be some insertionloss in any fiber optic connection.When the Agilent 8504B magnitudecalibration is executed, the loss atthe test port connection is effec-tively removed from the measure-ment of the calibration standard.However, when the DUT is con-nected to the test port, the insertionloss of this connection may be dif-ferent than the connection madeduring calibration. If the insertionloss is higher by 0.5 dB, subsequentreturn loss measurements will betwice this or 1 dB larger than theactual value. This is because boththe stimulus signal and the reflectedsignal will experience a 0.5 dBattenuation. Similarly, when the DUTconnection to the test port is betterthan the connection at calibration,return loss values will be worse thanactual by two times the insertionloss improvement.

It is therefore important that properconnector care and usage be observedto minimize the effects of connectorrepeatability.

Transmission path losses

Transmission path losses in the DUTaffect measurements in a way simi-lar to connector insertion lossrepeatability.

Large reflections can degrade themeasurement of other reflectionsfurther into the DUT. An examplewould be if a simple Fresnel air gapexisted in front of other reflections.3.5% of the energy is reflected atthe glass-to-air interface, 96.5%

travels to the air-to-glass interfacewhere again 3.5% of the energy isreflected. The net result is that only 93% of the stimulus signal (an 0.3 dB loss) reaches reflectionsbeyond the air gap. Similarly, only 93% of the reflected energywill reach the instrument detector.The result is similar to having an 0.3dB lossy splice, which would lead to0.6 dB measurement errors.

Measurement accuracy vs. position

A portion of the internal light path to the reference mirror in theAgilent 8504B is an open beam envi-ronment. As the reference mirror ismoved and the length of the openbeam path is increased, there will besome beam divergence. This beamdivergence can result in power varia-tion in the reference beam at thedetector. This phenomenon is veryrepeatable. Consequently, most ofthis effect is removed from the mea-surement as part of the factory cali-bration process.

The Agilent 8504B detection schemerequires that the reference mirrorideally moves at a constant velocity.However, there will be some velocityjitter.

The error due to the combination of the above two error sources isapproximately ±1 dB.

Sweep-to-sweep repeatability

Sweep-to-sweep repeatability is thesweep-to-sweep amplitude variationseen when measuring a known stablereflection. This error source is alsorelated to the mechanical movementof the interferometer mirror. It doesnot include the effects of noise whenmeasuring reflections near theinstrument noise noise floor. It isspecified to be less than ±0.5 dB.

Polarization sensitivity

The stimulus signal from the Agilent 8504B is randomly or onlypartially polarized. As the light trav-els to and from the DUT, its state ofpolarization will vary. Although theAgilent 8504B receiver is designedto be insensitive to the polarizationcharacteristics of the reflected light,there will be some uncertainty inamplitude measurement accuracydue to polarization, even if thesource stimulus is randomly polar-ized. This uncertainty is typicallyless than ±0.75 dB.

Calibration uncertainty

When a magnitude calibration isperformed with the Agilent 8504B,the instrument uses the knownreturn loss value of a calibrationstandard to remove many systematicerrors in subsequent measurements.The cables supplied with the Agilent8504B can be used as a reflectionstandard. The return loss value atthe open end is consistent within±0.1 dB. This uncertainty in theactual return loss of the calibrationstandard will result in uncertaintyin DUT measurements of approxi-mately ±0.1 dB.

To maintain this low uncertainty, thecalibration standards (typically theend of a fiber patch cord) must beclean and undamaged.

Chromatic dispersion effects

Part of the propagation path of thereference signal is in an open beam.In most cases, light traveling in theDUT path is all in fiber. When makingmeasurements at 1550 nm, theamount of chromatic dispersionexperienced by the light traveling inthe reference path will be less thanthat in the DUT path. This mismatchin dispersion results in a broadeningand subsequent drop in the peakvalue of the displayed reflection“impulse”.

The peak value will decrease mono-tonically as a function of the lengthof dispersion mismatch. This effect isconsistent and has been corrected outby the Agilent 8504B. The instrumentassumes a dispersion coefficient of17 ps/(nm*km). The result of thiscorrection leaves a residual error onthe order of ±0.3 dB.

The problem becomes difficult whenthe path to the DUT is both in fiberand an open beam. The effects arethen very difficult to remove from themeasurement, and subsequent uncer-tainties due to chromatic dispersioncan approach 5 dB. The user has theoption of disabling the internal disper-sion correction to facilitate his owncorrection methods.

Effects of source spectral width

The spectral width of the Agilent8504B source is approximately 55 nm.Another uncertainty component willexist if the DUT reflection character-istics vary over this spectral range.The level of uncertainty is dependenton the DUT characteristics.

BibliographyS. Newton, “Technology trends inoptical reflectometry”, PhotonicsSpectra, November 1991, pp. 118-126

H. Booster, H. Chou, “Higher Reso-lution for Backscatter Measurements”,Lasers and Optronics, October 1991,pp. 27-30

Positional Accuracy IssuesThe accuracy of the Agilent 8504B indetermining the relative location of reflections is based on its abilityto control and monitor the positionand velocity of the reference mirror.This uncertainty is less than 2% ofthe measurement span. To have thehighest accuracy, the narrowest spanthat includes the two events of inter-est should be used.

SummaryIn general, the individual error components are uncorrelated. Thetotal measurement uncertainty isdetermined with an "RSS" (RootSum Square) analysis, and not a linear summation.

Return loss (dB)14.6 14.65 14.7 14.75 14.8 14.85 14.9 14.95

654321R

elat

ive

occu

ranc

es

λ =1550 nmAvg =14.73 dBStd =±0.085 dB

Calibration Standard Repeatability

Return loss (dB)14.85 14.9 14.95 15.0 15.05 15.1 15.15 15.2

654321R

elat

ive

occu

ranc

es

λ =1300 nmAvg =15.003 dBStd =±0.065 dB

19

Agilent Technologies’Test and Measurement Support, Services, and AssistanceAgilent Technologies aims to maximize the value you receive, while minimizing your risk andproblems. We strive to ensure that you get the test and measurement capabilities you paid for andobtain the support you need. Our extensive support resources and services can help you choosethe right Agilent products for your applications and apply them successfully. Every instrumentand system we sell has a global warranty. Support is available for at least five years beyond theproduction life of the product. Two concepts underlie Agilent’s overall support policy: “OurPromise” and “Your Advantage.”

Our PromiseOur Promise means your Agilent test and measurement equipment will meet its advertisedperformance and functionality. When you are choosing new equipment, we will help you withproduct information, including realistic performance specifications and practical recommend-ations from experienced test engineers. When you use Agilent equipment, we can verify that itworks properly, help with product operation, and provide basic measurement assistance for theuse of specified capabilities, at no extra cost upon request. Many self-help tools are available.

Your AdvantageYour Advantage means that Agilent offers a wide range of additional expert test and measurementservices, which you can purchase according to your unique technical and business needs. Solveproblems efficiently and gain a competitive edge by contracting with us for calibration, extra-costupgrades, out-of-warranty repairs, and on-site education and training, as well as design, systemintegration, project management, and other professional engineering services. Experienced Agilentengineers and technicians worldwide can help you maximize your productivity, optimize thereturn on investment of your Agilent instruments and systems, and obtain dependablemeasurement accuracy for the life of those products.

By internet, phone, or fax, get assistance with all your test & measurement needs.

Online assistance:www.agilent.com/comms/lightwave

Phone or FaxUnited States:(tel) 1 800 452 4844

Canada:(tel) 1 877 894 4414(fax) (905) 282 6495

Europe:(tel) (31 20) 547 2323(fax) (31 20) 547 2390

Japan:(tel) (81) 426 56 7832(fax) (81) 426 56 7840

Latin America:(tel) (305) 269 7500(fax) (305) 269 7599

Australia:(tel) 1 800 629 485 (fax) (61 3) 9210 5947

New Zealand:(tel) 0 800 738 378 (fax) 64 4 495 8950

Asia Pacific:(tel) (852) 3197 7777(fax) (852) 2506 9284

Product specifications and descriptions in this document subject to change without notice.

Copyright © 1992 Agilent TechnologiesPrinted in USA January 1, 20015963-7191E