technical lumen maintenance testing of off‐grid lighting ......a short description of each method,...

Date:

Introduction

Consistent luminous flux over the lifetime of an LED

lightingproduct(referredtoaslumenmaintenance)isan importantperformancemetric. Lightingdesigners,manufacturers, importers, aswell as government and

non‐government organizations often require lumenmaintenance data to evaluate product quality. Forthosewhodonothavespecializedtrainingandaccess

to relatively high‐cost photometric equipment, thereexist simple, low‐cost devices that can accuratelymeasureLEDlumenmaintenance.Twosuchdevices,a

box‐photometer and a tube‐photometer, can beconstructed of widely available materials, using basictools,for$250‐$350.Inadditiontobeinginexpensive,

the tube‐ and box‐photometers occupy little space,and a minimally trained operator can make lumenmaintenancemeasurementsquicklywithbothdevices.

LEDLumenMaintenance

LEDsdonottendtofailcatastrophically.Instead,theyexperienceanirreversibledecreaseinlightoutputovertime called ‘lumen depreciation’. The inverse of

lumendepreciation,‘lumenmaintenance’isdefinedasthe percentage of the initial luminous flux outputremainingataspecifiedelapsedoperatingtime.1

Forexample,anLEDproductmaylose30%ofitsinitiallightoutputafterrunningfor2000hours.Itslumenmaintenancewouldthenbe70%at2000hours,alsowrittenasL70.Iftheproductinsteadretained90%ofitsinitiallightoutputafter6000hoursofoperation,itslumenmaintenancewouldbe“L90at6000hours”.

Modern off‐grid lighting technologies must becompetitive on the basis of cost‐per‐time of lighting

service if they are to be adopted by low‐incomeconsumers. Unfortunately, many low‐quality LED

©LightingAfrica–April2012

TechnicalNotesIssue8April2012

LumenMaintenanceTestingofOff‐gridLightingProductsThisTechnicalNotecoversmethodsformeasuringlumenmaintenanceofLEDbasedoff‐gridlightingproducts,focusingonlow‐costmethodstoevaluateaproduct’sabilitytoprovidelong‐termlightingservice.TheInformationcontainedinthisarticlebuildsonpreviousTechnicalNotes,seealso:http://www.lightingafrica.org/resources/briefing‐notes.html

products exhibit such rapid lumen depreciation that

the potential economic savings are never achieved.With this in mind, the Lighting Africa Quality TestMethod (LA‐QTM) and Initial Screening Method (LA‐

ISM) include an off‐grid lighting‐specific method formeasuring lumen maintenance that is derived fromexisting standards and internationally accepted

methods. Examples of lumen maintenance plots for“stable”andrapidlydepreciatinglightingproductsareshowninFigures1and2.

Figure1.Lumen maintenance plot for a product that meets Lighting Africa minimum standards.

Figure2.Lumen maintenance plot for a product that fails Lighting Africa minimum standards.

LumenMaintenanceTestingOff‐gridLightingProductsIssue8April2012

ReferenceTestMethods

The test method established in LM‐80‐08 by the

Illuminating Engineering Society (IES) serves as theprimarymethodologicalreferencefordeterminingthelumenmaintenanceofindividualLEDcomponents.LM‐

80‐08specifiesthatanLED,LEDarray,orLEDmodulemust be operated for at least 6,000 hours,with datacollectionevery1,000hours,resultinginaminimumof

sixmeasurementsoverthecourseofthetest. Thesemeasurements are typically conducted with anintegrating sphere‐spectrophotometer system, as

described by the International Commission onIllumination(CIE)technicalreportonthemeasurementof luminous flux (CIE 84‐1989). LM‐80‐08 states that

LEDs are to be driven at a constant current that isspecified by the LED manufacturer, and must beregulated to within±3% of the rated value over thetesting period. The ambient air temperature is

maintained at 25oC ±2oC and airflow must be

minimizedinordertoreducetheeffectsofconvectiveheat transfer away from the LED package. A lumenmaintenance test report must include all of the

operating conditionsand thepercent change in initialluminousfluxateachmeasurementinterval.

LM‐80‐08 is an accurate method for measuring thechange in LED luminous flux for an LED source over

time. Alterations to the LM‐80‐08 procedure havebeen incorporated into the LA‐QTM and LA‐ISM inorder to allow and simplify testing of LED‐based off‐

grid lighting products. These changes help distinguishbetween products that experience rapid lumendepreciation and those that are able to provide

thousands of hours of lighting service while reducingthetime,expense,andcomplexityoflifetimetesting.

LumenMaintenanceTestingProcedure

The LightingAfricaprogramspecifies that theoff‐gridlightingproductbecontinuouslyoperatedforaperiod

of500hours forLA‐ISMand2,000hours forLA‐QTM.

The lightoutput ismeasuredatregular intervalsusing

a fixed‐geometrymeasurement cavity under constantconditions.2 The light output is quantified in units oflumens (for luminous flux measurements) or lux (for

illuminancemeasurements),dependingonthetypeoftestingdeviceused.

Foreachlightoutputmeasurement,thefollowingmust

bereported:

• Relative light output (initial measurement dividedbythemeasurementataspecifictimeinterval)

• Operatingtimeofthelightsource• Ambientairtemperature• Currentandvoltagedeliveredtothelightsource

Measurementfrequency

The LA‐QTM requires at least 17 light outputmeasurements are conducted over the course of the

test. The LA‐ISM requires at least 6 light outputmeasurements.Theinitialmeasurementistakenaftera 20minutewarmup period in order for the light to

reach steady state – if for any reason the power isinterrupted during the warm up period, a new 20minute warm up period must be given. The initial

measurement should be repeated at least once tocheck for an erroneous measurement. The secondmeasurement should be made 24 hours later to

determine if the product is likely to experienceextremely rapid lumen depreciation. If the relative

luminous flux at the second measurement isapproximately 95% or less, then the measurementfrequencyneedstobeincreasedtocapturethetimeat

which L70 is reached. The LA‐QTM indicates that thenext five light outputmeasurements should be takenat 48‐hour intervals, which are then followed by

measurements at one‐week intervals. LA‐ISMrecommendsthatmeasurementsbeconductedat100‐hour intervals. Discretion must be exercised to

determine the appropriate measurement frequency,according to the resolution desired as well asavailabilityofequipmentandoperators.

©LightingAfrica‐April2012

Orientationofthelightingproduct

Overtheperiodofcontinuousillumination,thedeviceunder test should be oriented as it is intended fortypical use, in order to replicate the heat transfer

occurringduringnormaloperation.

RecommendedTestingDevices

LightingAfricarecommendsthreedifferentdevicesformeasuring lumen maintenance of an off‐grid lightingproduct,inorderofdecreasingcostandcomplexity:

• Integratingsphere‐spectrometersystem• Self‐builtbox‐photometer

• Self‐builttube‐photometer

Below,isashortdescriptionofeachmethod,includingrecommendationsandadviceforproperapplication.

IntegratingSphere

An integrating sphere is a hollow sphere whoseinternal surface is a diffuse reflector that spatiallyintegratesradiantflux.3Whileanintegratingsphereis

the most complex and expensive option for

determining lumen maintenance, it is also the mostversatile and can be used for several otherphotometric measurements including: radiant and

luminous flux, laser power, reflectance andtransmittanceofmaterials.AtypicalintegratingsphereisshownbelowinFigure3.

Figure3Integratingsphere(http://www.optronik.de/plm‐is‐its10.shtml).



Ambientconditions

The ambient air temperaturemust bemaintained at25 °C ± 3 °C, with minimal airflow. LEDs are

temperature sensitive and large temperaturefluctuationsorexcessiveairflowcanresultinerrorsinlumenmaintenancemeasurements.

PowerSupply

The device under test (DuT) battery is replaced by aDC power supply set to deliver constant voltage

(measuredatthedevice)equaltothenominalbatteryvoltage.Thisvoltagemustberegulatedtowithin±3% for the duration of the testing period and measured with an accuracy of ± 0.2% at each photometric measurement. The power supply should have aresolution of at least 0.01V and 0.001A, and becapableofremotevoltagesensing(usedwitha4‐wire

measurement technique) in order to compensate forresistanceandvoltagedropsinwiresandconnectionsto the lighting device. For power supplies without

remote sensing, the voltage can be measured veryclosetothelightingdeviceundertesttoconfirmthatthe voltage input is equal to the nominal battery

voltage.Paired,multi‐strandwireofthetypetypicallyusedforspeakers (18‐22AWGor0.258‐0.823mm2)is acceptable for connecting the lighting product to

thepowersupply–heaviergaugewire (>0.823mm2)ispreferred.

During testing, test personnel should be aware of

nominal drive current and voltage levels for DuT’s.Anydiscrepancybetweenexpectedandactualcurrentor voltage values should be noted, and the test

apparatus should be inspected for possible causes.Sincethetestingprovidesaconstantvoltage,andnota constant current, it is not usually possible to

regulate the power being delivered to the LEDsource(s). The DuT electronics regulate this power,and there can be significant drift in this power

regulation that can affect the light output of thedevice.



An integrating sphere is advantageous for use in

lumen maintenance testing because it is lesssusceptiblethanothercavitiestomeasurementerrorassociated with changes in the location and

orientation of the light source within the cavity.Nonetheless, when conducting lumen maintenancemeasurements,thelightingproductshouldbelocated

inexactly thesamealignmentandorientationwithinthespheretominimizethepotentialforthiserror.

Thedrawbacksofanintegratingsphere,especiallyfor

users who do not specialize in photometry, includethe relatively high cost of a sphere and associatedcomponents, the need for frequent calibration,

relatively time‐intensivemeasurements, theneed forspecializedtrainingtooperatethesystemandanalyzethedata, and relatively large spatial requirements (a

1mdiameter integratingsphere,sufficient for testingmostoff‐gridlightingproducts,typicallyoccupies4m2offloorspace).

Accurate luminous fluxmeasurementsmadewith anintegrating sphere require calibration of the systemaccording toa ‘standard lamp’of known radiant flux

andcorrectionofthelumenmeasurementtoaccountforself‐absorptionofthebodyofthelightingproduct.Themethods used to calibrate and correct luminous

fluxmeasurementsmadewith an integrating spherearenotcoveredhere,astheyrequireamuchmorein‐depth explanation, which will be covered in a

forthcomingTechnicalNote.3

Box‐Photometer

A box‐photometer is an integrating photometeremployinganarbitrarilyshaped,hollowboxorcavitythatcanbeusedtocomparetheluminousfluxoflightsources of the same type. The box‐photometer isuseful and convenient for measuring the relativechangeinlightoutputofalightsource,asrequiredtoevaluate lumenmaintenance. Aninteriorviewandathreedimensionalrenderingofabox‐photometerareshowninFigure4.4

A typical low‐cost box‐photometer is a hollow box

constructedofwoodwiththeinnersurfacescoveredinwhitemattepaint.Multiplecoatsofpaintimprovetheuniformity and reflection of the walls. The finished

interior surface is intended to reflect lightasdiffuselyand homogeneously as possible. Caremust be takento apply the paint such that the surface texture is

smooth. Midway up one corner of the box is a portthat allows an illuminancemeter sensor to “see” intothe box. A baffle that is constructed of a thin, rigid

material(0.5cm‐thickplywoodisappropriate)isplacedin front of the sensor port so that light emitted from

the device under test cannot shine directly on thesensor.Thelightsourcetobemeasuredisplacedonastand that is located in the center of the box. When

closed, the box must not allow any ambient light toenter.RefertoAppendixAfordetailedplansforabox‐photometer that is suitable for lumen maintenance

testingoftypicaloff‐gridlightingproducts.

Relative light output measurements with a box‐photometer do not require calibration or device‐

specific corrections to deliver quick and accurateresults. Sincenocalibrationor correction is required,the light source is simply placed in the center of the

box and the illuminance measured by the lux/foot‐candle meter is recorded. It is extremely importantthat for each illuminance measurement, the light is

placed in the box in exactly the same location, withexactly the same orientation. Even the slightestadjustment in the location or orientation of the light

sourcebetweenmeasurementscanresultinsignificanterrorintherelativelightoutputmeasurements.

Figure4.Box‐photometer(left:Interiorview,right:3‐Dview)



Two approaches can be taken when using the box‐photometer to conduct lumen maintenance

measurements. Theboxcanbededicatedtoasinglelight source, or the device can be used to measuremultiplelightsources.

Dedicatedbox‐photometer

A single product sample remains in the box for theentire durationof the test. Thebenefit of dedicatingthe box to a single light is that the product can be

fixed in the box, thereby reducing the possibility ofmeasurement error due to changes in alignment. Inaddition, ifa luxmeterwithdata loggingcapability is

used, then the illuminance can be continuouslymonitoredandrecordedwithminimaloperatoreffort.Care should be taken to check that the light source

does not build up heatwithin the cavity, as this canaffect both the source and the sensor.Most off‐gridproducts,however,willnotdissipateenoughpowerto

havesignificantthermaleffectsinthebox.

Box‐photometerformultiplesamples

Theboxisusedtomeasuremultiplelightingproductsthat are operated outside of the box between

measurements. Using the box‐photometer tointermittently measure multiple samples is morecommonthanthe“dedicated”approach,as itgreatly

increases the testing throughput of the box‐photometer.Inthiscase,thetestoperatormusttakeextreme care to place the light in exactly the same

location and orientation for each subsequentmeasurement. Several methods can be used toensure replicate product placement within the box,

including:• using a photo for reference that shows the

placementofthelight• usinga custom jigwithin thebox thatholds the

lightinexactlythesameorientation• markson theproductand test stand toassist in

placementandalignment

Measuringluminousfluxwithabox‐photometer

Luminousfluxisusuallymeasuredwithanintegratingsphere in accordancewith CIE 84‐1989. However, ifan integrating sphere is not available, a box‐

photometercanbeusedformeasuringluminousflux.Inpractice,however,measuring lumenoutputwithabox‐photometer is not recommended since it is

relatively cumbersome and inherently more errorprone than an integrating sphere‐spectrometersystem. Modern lighting laboratories generally rely

onintegratingspherestomakelumenmeasurementsonaccountoftheeaseofuse,measurementaccuracy,andminimalimpactofspatialirregularities.

Tube‐Photometer

A tube‐photometer is a simple, hand‐made deviceusedtomeasuretherelativechangeinilluminanceof

a light source over time. The apparatus consists oftubewithacapatoneend.Theendcapfitssnuglyonthe tube such that it holds the illuminance meter

sensorinafixedpositionandrestrictsstraylightfromenteringthetube.Lightemittedbyalampundertestis directed into the open end of the tube and the

illuminance incidenton the sensorhead ismeasuredbytheluxmeter. Adiagramofthetube‐photometerisshowninFigure5.

Figure5.Drawingofatube‐photometer,indicatingthebasicdevicecomponents.

Materialsforthetubeandendcapcanbeprocuredatvery low cost, often free of cost. Cardboard tubes

that are commonly used in rolls of fabric, paper andplastic sheets tend to be widely available andinexpensive, if not free. Alternatively, inexpensive

plasticpipecanbeused.Atypicaltubediameteris6



cm and an appropriate length for the tube‐

photometer is 0.5 m, although these dimensions donotneedtobestrictlyadheredto.Ideally,theendcapismachinedfromarigidmaterial likewoodorplastic

that fits snugly into the tube and mates with thephotometer sensor such that the sensor is fixed inplace. Research has shown that if precisemachining

equipmentisnotavailable,anacceptableendcapcanbe simply constructed fromcardboard andpackagingtape.5 Refer to Appendix B for instructions for

constructingtheendcap.

Unlike the integrating sphere and box‐photometer,the luminous flux of a light source cannot bedetermined by the tube‐photometer. The tube‐

photometer is used solely for evaluating the lumendepreciation characteristics of a light source. Byplacingthedeviceundertestflushwiththeopenend

of the tube, the orientation anddistance of the lightsource from the sensor head can be replicated for eachilluminancereadingtakenthroughoutthelumen

maintenance test. As long as the same orientationbetweenthelightsourceandsensorismaintainedforeach reading, an accurate measurement of relative

changeinluminousfluxcanbeachieved. Similar to the box‐photometer, the tube‐photometercan be dedicated to measuring the lumenmaintenance of a single light source, or used to

measuremultipletestsamples.

Dedicatedtube‐photometer

Atube‐photometerisdedicatedtoasinglelightthatisfixedtotheopenendofthetube.Extremecaremust

betakentoaffixthetestproducttothetubesuchthatrepositioningdoesnotoccurbetweenmeasurements.

Tube‐photometerformultipletestsamples

Asingletube‐photometerisusedtomeasuremultiple

product samples that are not attached to the open

endofthetube.Rather,thetestoperatorholdsthelight source flush to the open end of the tube.Maintainingthe lightsourceflushwiththetube,the

operator makes slight adjustments to the anglebetweenthelightsourceandthetube,searchingforthe maximum illuminance measured by the light

meter.Thisapproachrequiresameterthathastheability to record and display the maximumilluminance reading. The test operator continues

“searching” until the maximum illuminance readingdisplayedon the lightmeterno longer increases foranyorientationof thedeviceunder test. Forhighly

directional light sources like small task lights andtorches (flashlights), all of the emitted light can beshined into the open end of the tube. Wider

distribution light sources (lanterns/ambient lights)mayrequiretheoperatortoselectaspecificregionin

whichtosearchforthemaximumilluminance.

When using this approach, the test operator mustsearch diligently for the maximum value, onlyrecordingthemax luxvaluewhenallorientationsof

the light source have been evaluated. The operatormustalsopreventstray light fromenteringthetubeby covering the light source and tube opening with

anopaque clothor by taking themeasurement in adarkroom.

Box‐andTube‐PhotometerReliability

An analysis of the box‐ and tube‐photometerconfirms that the devices can be used to accurately

measuretherelativechangeinlightoutputoftypicaloff‐grid lighting products.5 The study compares

relative light output measurements made with thebox and tube to those made with an integratingsphere. Results indicate that these methods are

acceptable substitutes for an integrating spherewhenconductinglumenmaintenancemeasurementsof light sources with varying physical shapes, sizes,

lumenoutputs,andspatiallightdistributions.

The analysis does warn that errors can result fromimproper use of the tube‐ and box‐photometers.Specifically, test operators must take care to avoid

changes in the light source location and orientationwithin the box. When testing multiple samples, theoperator must be aware of error due to false

identification of the maximum illuminance. Theseerrorscanbeavoidedbyrepeatingthemeasurementprocedure and training test operators in the proper

useoftheequipment.

Absolutevs.RelativeTesting

Relative testing measures the change in a measuredtest quantity. For lumen maintenance testing, the

procedures described herein measure the relativechangeinlightoutputofalightsourceoverasettimeperiod. Equipment calibrations are less stringent for

relativeteststhanforabsolutetests.

Absolute testing, as the name implies, measuresabsolute quantities (lumens, lux, etc. represent

quantitiesofvisiblelight).Absolutetestresultsrequirecalibrated instruments that have been properlychecked and adjusted for accuracy. Ultimately, a

properlycalibratedinstrumentwillhaveitscalibrationperformance ‘traceable’ to a proper standardmaintainedbyaprofessionalstandardsorganization.

For example, an illuminance meter may be out of

calibrationand ‘off’by5 lux.Absolutemeasurementswill thus be inaccurate. However, this same meter

maybeusedtomeasurethechangeinlux(asduringalumenmaintenancetest)aslongasitisstableandnochange has occurred to the meter during the test

period(so itmustremain5 lux ‘off’attheendofthetest, too). In thisway, relative testing can cancel theerrorincalibrationofthetestequipment.

Conclusion

Lumen maintenance is a critical performance metric

thatmustbemeasuredwhenevaluatingthequalityofLED‐based lighting products. The box‐ and tube‐photometeraresimpleandaffordabledevicesthatcan

be constructed and used by non‐specializedtechnicianstoaccuratelymeasurelumenmaintenance.Relative lumen maintenance measurements made

with the box and tube require less calibration thanabsolutemeasurements.Theaccuracymeasurementsmade with these devices, however, is highly

dependent on stable operating conditions as well asconsistentlocationandorientationofthelightsource.

References

1. For more information about the LED lumendepreciation and lifetime, refer to Issue 2 of

LightingAfricaBriefing Notes: LED LumenDepreciationandLifetime

2. Refer to the Lighting Africa Quality Test Method

(LA‐QTM)fora complete description of the Long‐Term LumenDegradationTest, available fordownloadat www.lightingafrica.org

3. Labsphere. (2008). A Guide to Integrating SphereTheory and Applications. Retrieved January 2011.http://www.labsphere.com/uploads/technicalguid

es/a‐guide‐to‐integrating‐sphere‐theory‐and‐applications.pdf

4. Basic instructions for construction of a box‐

photometer are included in Lighting AfricaQualityTestMethod(LA‐ QTM) available fordownloadatwww.lightingafrica.org

5. Carlsen, C. (2011). Analysis of Low‐cost testingmethods for LED lumen maintenance of off‐gridlightingproducts. Humboldt State Universityhttp://humboldt‐dspace.calstate.edu/handle/2148/766



APPENDIXA:Box‐PhotometerConstructionLumenMaintenanceTestingOff‐gridLightingProducts


APPENDIXA:Box‐PhotometerConstructionLumenMaintenanceTestingOff‐gridLightingProductsIssue7January2012

©LightingAfrica–April2012

TUBE‐PHOTOMETERENDCAP

Atube‐photometerendcapisrequiredtoholdthelightmetersensorinplaceandtorestrictambientlightfromenteringintothetubefromoneend.Theendcapispreferablymadeofarigidmaterial(typicallywood),butanequally

acceptableendcapcanbefashionedfromcardboardandadhesivetapeifprecisionwoodworkingtoolsarenotavailable.Alatheistypicallyrequiredtofabricateawoodendcap.Thecardboardendcapcanbecraftedusingautilityknifeandadhesivetape.Forbothtypesofendcap,greatcaremustbetakentocutthecaptotheprecise

measurementsofthelightmetersensordiameterandtubediametertoensuretightfits.Photosofawoodandcardboardendcapforthetube‐photometerareincludedbelowasareference.

APPENDIXB:Tube‐PhotometerConstructionLumenMaintenanceTestingOff‐gridLightingProducts


technical lumen maintenance testing of off‐grid lighting ......a short description of each method,...

Documents