diodes and light

11
1 DIODES AND LIGHT LAB 6 INTRO: INTRODUCTION TO DIODES, LEDS, PHOTODIODES, AND TRANSIMPEDANCE AMPLIFIERS GOALS In this lab, you will design and build a photometer (optical detector) based on a silicon photodiode and a current-to-voltage amplifier whose output is proportional to the intensity of incident light. First, you will use it to measure the room light intensity. Then you will set up and investigate an optical communication link in which the transmitter is a light emitting diode (LED) and the receiver is your photodiode detector. Proficiency with new equipment: o Diodes, LEDs, photodiodes, and transimpedance amplifiers Modeling the physical system: o Model the sensitivity of a photodiode o Develop a model of a photometer measurement system Applications: o Build a calibrated photometer o Build an optical communications link DEFINITIONS Intensity – power per unit area, unit: W/cm 2 Luminous intensity – power perceived by the eye emitted per unit solid angle, unit: mcd (millicandela) Luminous flux – total power emitted as perceived by the eye, unit: lumen DIODES - GENERAL

Upload: others

Post on 16-Nov-2021

1 views

Category:

Documents


0 download

TRANSCRIPT

1

DIODESANDLIGHTLAB6INTRO:INTRODUCTIONTODIODES,LEDS,PHOTODIODES,ANDTRANSIMPEDANCEAMPLIFIERS

GOALSInthislab,youwilldesignandbuildaphotometer(opticaldetector)basedonasiliconphotodiodeandacurrent-to-voltageamplifierwhoseoutputisproportionaltotheintensityofincidentlight.First,youwilluseittomeasuretheroomlightintensity.Thenyouwillsetupandinvestigateanopticalcommunicationlinkinwhichthetransmitterisalightemittingdiode(LED)andthereceiverisyourphotodiodedetector.

Proficiencywithnewequipment:

o Diodes,LEDs,photodiodes,andtransimpedanceamplifiers

Modelingthephysicalsystem:

o Modelthesensitivityofaphotodiodeo Developamodelofaphotometermeasurementsystem

Applications:

o Buildacalibratedphotometero Buildanopticalcommunicationslink

DEFINITIONSIntensity–powerperunitarea,unit:W/cm2Luminousintensity–powerperceivedbytheeyeemittedperunitsolidangle,unit:mcd(millicandela)Luminousflux–totalpoweremittedasperceivedbytheeye,unit:lumen

DIODES-GENERAL

2

Figure1:DiodeCharacteristics

Adiodeisasemiconductordevicethathastwoterminals,ananodeandacathode.Theseterminalsarerepresentedby(-)forthecathodeand(+)fortheanodeonthediagraminFigure1.Diodesaremadeoutofn-dopedandp-dopedsemiconductors.Foragoodexampleofhowthesework,see:https://phet.colorado.edu/en/simulation/semiconductor

The fundamental property of a diode is its tendency to conduct electriccurrentin only onedirection.Thediodecanbeoperatedinthreeregions.Thefirstregion,calledforwardbiased,iswhenthecathodehasahigherpotentialrelativetotheanode.Oncethispotentialdifferenceisgreaterthansomethreshold voltage (~0.6 V for silicon diodes) the diode conducts with almost zero resistance and has aconstant0.6Vdropacrossit.Thesecondregionofoperationiswherethecathodehasalowerpotentialrelativetotheanode.Thisregion iscalledreversedbiasedandessentiallynocurrentcanflow.Finally, ifthe potential difference from the cathode to the anode is negative and larger than some breakdownvoltage, thediodewill again conduct.Regulardiodesarenotused in thebreakdown region,but Zenerdiodesareusedforthispurpose.

LIGHTEMITTINGDIODE(LED)TheHLMP-C625lightemittingdiodeactselectricallylikeanydiode.However,unlikeregulardiodesitalsoemitslightwhenforward-biasedduetodirectradiativerecombinationofelectronsandholes.Theforwardvoltagedropisabout1.9Vratherthan0.6VbecausetheLEDismadeofAlInGaPinsteadofsilicon. TodescribetheoutputofalightsourcelikeourLED,it ishelpfultointroducethenotionofsolidangle.Consideratransparentsphereofradiusr,andsupposethatanareaAonthesurfaceofthesphereispaintedwhite(Figure2).Wethensaythatthewhited-outregionsubtendsasolidangleofΩsteradians(sr),whereΩ=A/r2.Accordingtothisdefinition,thewholespheresubtendsasolidangleof4πsr.Onesteradianisanareaofr2,justasoneradianisanarcoflengthr.

Figure2:SolidAngledepiction. The concept of solid angle is essential in separating the two units in which light is customarilymeasured.Boththelumenandthecandelaoriginatedinthe18thcenturywhentheeyewastheprimarydetectorofelectromagneticradiation. Thelumen(lm)isameasureofthetotallightpoweremittedbyasource.Youmightthenexpectthat there is a conversion factor between lumens andWatts, and you would be right: its value is 683lm/Watt. However,thingsareabitmorecomplicatedbecausethisconversionfactor isonlycorrectforlightwithawavelengthof555nm,theyellow-greencolorthatoureyesaremostsensitiveto.Forothercolors the conversion factor ismultiplied by a dimensionless number y(λ) called the photopic standardluminosityfunction.Aplotofy(λ) isshowninFigure3andatabulationofthisfunctioncanbefoundonthecoursewebsiteundertheUsefulDocsSectionatthebottom:photopicstandardluminosityfunction.Thepointofthisisthattwosourcesdescribedbythesamenumberoflumens(thesame“luminousflux”)will have the same subjectivebrightness toahumanobserver. This kindof color correctedunit is very

3

helpfulifyouwanttodesignacontrolpanelwithlotsofcoloredlights,andyouwantthemalltohavethesameperceivedbrightness. Tosummarize, if the luminous fluxofyoursource isdescribedasF lumens,thenyouconvertthistoWattsusingthisformula:

( )

( )( )lm

6831

W Fy

Fλ⋅

= (1)

NoticethatmoreWattsarerequiredforagivenluminousfluxasthecolorgetsfartherandfartherawayfromyellow-green,tomakeupforthedecliningsensitivityoftheeye.Note:Thisisreallyjustaconversionfromoneunittoanother. However, this is not the whole story for describing light sources, because the amount of lightemittedvarieswithdirection,andhowmuchlightweinterceptinagivendirectionwilldependuponhowmuchsolidangleourdetectorcovers.Thusweneedameasureoflightpowerpersolidangle,andthisunitiscalledthecandela,equaltoonelumen/sr.Alightsourcethatemitsonecandelaineverydirectionemitsatotalof4πlumen,sincethereare4πsr inthewholesphere. Thequantitymeasuredbythecandelaiscalledthe“luminousintensity”. IfyoulookatthedatasheetforourHLMP-C625LEDyouwillseethatitusestheunit“mcd”ormillicandelatodescribethebrightness.ThevaluesgivenareforlightemittedalongtheaxisoftheLED.ForotherdirectionsseetheSpatialDistributiongraphinthedatasheet.BydividingEqn.4abovebythesolidanglewecanrewriteitasarelationbetweentheluminousintensityJinmcdandthepowerperunitsolidangle:

( ) ( )mcd)(683

1mW/str J

yJ

λ⋅= (2)

Again:Notethatthisisreallyjustaconversionfromoneunittoanother.Suppose now we place our photodiode a distance R from the LED, and we want to find the intensityN(mW/cm2) at the photodiode. We first find J inmillicandela on the LEDdata sheet.We then convertJ(mcd) to J(mW/sr), using Equation 2 and y(λ) for the appropriate wavelength. (For our LED, y(650nm)=0.107.)FinallywedivideJ(mW/sr)byR2togetN(mW/cm2).

𝑁𝑚𝑊𝑐𝑚! =

1683 𝑦(𝜆)

1𝑅!(𝑐𝑚!)

𝐽 (𝑚𝑐𝑑)

Figure3:y(λ )asafunctionofwavelength.

4

PHOTODIODEThe PD204 photodiode used in this experiment is a p-intrinsic-n (PIN) silicon diode operated in reversebias.AsketchofthephotodiodestructureisshowninFigure4.Theverythinp-typeconductinglayeractsas a window to admit light into the crystal. The reverse bias voltagemaintains a strong electric fieldthroughouttheintrinsicregionforminganextendeddepletionlayer.Thedepletionlayershouldbethickerthantheabsorptionlengthforphotonsinsiliconinordertomaximizetheefficiency.Anyincidentphotonwhoseenergyexceedstheband-gapenergyisabsorbedtoproduceanelectron-holepairbyphotoelectricexcitationofavalenceelectronintotheconductionband.Thechargecarriersaresweptoutofthecrystalbytheinternalelectricfieldtoappearasaphotocurrentattheterminals.Thephotocurrentisproportionaltolightintensityoverarangeofmorethan6ordersofmagnitude.

+

incident light

+ –

p-type layer

n-type layer

intrinsic region

photocurrent –

electron-hole pair

Figure4:DiagramandschematicsymbolforaPINsiliconphotodiode.

The photodiode sensitivity Sλ (in units of µA/(mW/cm2)) is defined as the photocurrent per unit lightintensity incident on the photodiode. It is a function of the lightwavelengthλ. For light intensity (inmW/cm2)thephotocurrentI(inµA)isgivenby

I = SλN. (3)

The sensitivity at anywavelengthλ is given on the data sheet in terms of the peak sensitivity R940nm at940nmtimesacorrectionfactorcalledtherelativespectralsensitivity,orRSR:

Figure5showstheRSRfromthePD204datasheet.Youcanseethemaximumsensitivityisaround940nm.

).(nm 940 λλ RSRRS =

5

Figure5:PD204RelativeSpectralSensitivity

CURRENT-TO-VOLTAGEAMPLIFIER(TRANSIMPEDANCEAMPLIFIER) Inanordinaryinvertingamplifiertheinputvoltageisappliedtoaresistor,andtheamplifiergeneratesanoutputvoltageinresponsetothecurrentthatflowsthroughtheinputresistortothevirtualgroundatthenegative op-amp input. A current-to-voltage amplifier (Fig. 6) is an inverting amplifier with the inputcurrentIinsuppliedbythephotodiodeandappliedtotheinvertingop-ampinput.Sincenocurrentflowsinto theop-amp input, theoutput voltagemust beVout = –IinRF. The ideal (GoldenRules result) low-frequencygainofacurrent-to-voltageamplifieris

𝐺 =𝑉!"#𝐼!"

= −𝑅! Thisgainhastheunitsofimpedancei.e,Ohms,anditisoftencalledatransimpedanceortransimpedancegain. The current-to-voltage amplifier is sometimes called a trans-impedance amplifier. This type ofamplifier is very common in research labs, and is used to amplify the current fromphotodiodes, photomultipliertubes,iondetectors,etc.

Figure6:Photodetectorcircuit.

Everlight Electronics Co., Ltd. http:\\www.everlight.com Rev 1.1 Page: 4 of 7

Device NoΚDPD-020-033 Prepared dateΚ07-22-2004 Prepared byΚJaine Tsai

0-25

50

100

150

250 50

200

75 85 1000

0.2

0.4

0.6

0.8

1.0Ta=25

100 300 500 700 900 1100 1300

OC

120

10

100

40 60

VR=10V

80 100

1000

1.0

15

5

0.50

10

20

VR=5V

1.52

3.0

PD204-6C/L3 Typical Electro-Optical Characteristics Curves Fig.1 Power Dissipation vs. Fig.2 Spectral Sensitivity

Ambient Temperature

Fig.3 Dark Current vs. Fig. 4 Reverse Light Current vs. Ambient Temperature Ee

6

In our photometer circuit the current Iin flows through the back-biased photodiode when it isilluminated (it flows out of the op-amp negative input node and the resulting Vout is positive). ThefeedbackcapacitorCFenhancesstability i.e., ithelpstoavoidspontaneousoscillationsof theop-ampbyreducingthebandwidthoftheamplifier(justlikeanactivelow-passfilter).

LF356PINOUTANDSCHEMATIC

Allop-ampcircuitsstartoutbymakingthebasicpowerconnections.Op-ampsareactivecomponents,whichmeanstheyneedexternalpowertofunctionunlikepassivecomponentssuchasresistors.

Figure7.LF356pin-outandschematic.

USEFULREADINGS

1. Data sheets for the PD204-6C/L3 silicon photodiode and the HLMP-C625 AlInGaP light emittingdiodeareavailableatthecoursewebsite.

2. Optionalreadingondiodes:FCChapter4,particularlySections4.18&4.19

3. Forgeneralbackgroundonoptoelectronics,seeH&HSection9.10.

LABPREPACTIVITIES

AnswerthefollowingquestionsusingMathematica.BringanelectroniccopyofyourMathematicanotebooktolab,preferablyonyourownlaptop.

7

Question0 Introducingdiodesa. Consider the circuit in Fig. 7b. If the diode is ideal, predict the behavior of the current I

(consider clockwise current tobepositive) andVout as a functionof the source voltageVin.MakeatableofthepredictedIandVoutforthefollowingvaluesofVin:−2V,0V,+0.5V,+0.6V,+0.7V,+1V,+2V,+4V.

b. Now, consider the case where Vin is a sine wave with a 3V amplitude (6V peak-to-peak)centeredonV=0.PredictVoutandIasafunctionoftime,andsketchyourpredictionsover2-3periods.

Question1

RoomLightPhotometer

a. Estimate the sensitivity, Sλ, in units ofµA/(mW/cm2) of the PD204 photodiode tothefluorescentlightsinthelab.Thepeaksensitivityat940nmis10µA/(mW/cm2).Seethephotodiodedatasheetpostedonthecoursewebsiteor theplotabovetodetermine the RSR at the appropriate wavelength. The mean wavelength of thewhitefluorescentlightsisabout550nm.Considerif550nmmakessense…isthereawayyoucouldestimatethemean?

b. WhatintensityofwhitelightinmW/cm2doyouexpectisincidentonyourphotodiodeonthelabbenchwhenitisfacingupwardsi.e.,towardsthefluorescentlights?Eachfluorescentlighttube produces approximately 4W of visible light. You can assume that half of it (~2W) isemitteddownwardsinto2πsr.

c. Besuretostateyourassumptionsexplicitlyforpart(b).Howmanybulbsdidyoumodelandatwhatdistancefromthedetectorwaseachbulb?Whatwavelengthareyouassuming?Etc.Stateatleastthreekeyassumptionsinyourmodelprediction.

Question2

Trans-impedanceAmplifiera. For thecurrent-to-voltageamplifier inFig.6,chooseavalue for the feedbackresistorRFso

thatanincidentwhite-lightintensityNof1.0mW/cm2producesanoutputvoltageof10V.b. The small feedback capacitor CF is used to suppress spontaneous oscillations. The trans-

impedance gainof the amplifier at anyparticular frequency is -ZF,where ZF is the effectiveimpedanceoftheparallelRFCFcircuit.Thegainrollsoffathighfrequencieswithabandwidthof fB=1/(2πRFCF), just like any low-pass filter. The bandwidth will suffer if CF is too large.WhatisthebandwidthfBifCF=10pF?

c. Whatarethedcvaluesofthevoltagesatthe+and–inputsandattheoutputoftheop-ampfor zero light on the photodiode? Consider the ideal situationwhere the diode allows zerocurrent when there is no light hitting it. (In the full model, ~10 nA of dark current flowsthroughthediodeevenwithnolightonit.)

d. Whatwould the voltages be if the photodiode leadswere accidentally reversed tomake itforwardbiased?Hint:Inthiscasethediodesactslikeashortcircuitwitha0.6Vdropacrossit.

8

Question3 OpticalCommunicationLinka. AssumewehaveanHLMP-C625LEDbeingrunwithacurrentof20mAasinFig.9(left).Use

thetypicalvalueofluminousintensityat20mA(SeetheLEDdatasheetonthecoursewebsitetofindthetypicalluminousintensity).ComputetheintensityN(inunitsofmW/cm2)incidenton a detector place R cm away from the center of the transmitted light beam. Plot theintensityonthephotodiodeasafunctionofRforreasonabledistancesof1-50cm.

b. Plot the expected output voltage from the optical receiver as a function of the distancebetween thedetectorand theLED.Remember to recalculate the sensitivityof thedetectorforthewavelengthoflightfromtheLED,whichis650nmfortheonesweuseinlab.

c. Todrive the transmitter, the functiongeneratorwill beadjusted toproducea squarewavewithahighlevelof10V(unloaded)andalowlevelof0V.Thehigh-levelshouldgive20mAforwardcurrentintheLED,andthelowlevelshouldgive0mA.FindthevalueoftheseriesresistorRsthatgivesthecorrectcurrent.ThedatasheetliststheLEDforwardvoltagedropat20mAtobe1.9V(insteadof0.6Vforsilicondiodes).Inaddition,donotforgettomodelthe function generator as an ideal voltage supplywith a series output impedance of 50Ω,which is the true output impedance regardless of the “impedance setting” on the functiongenerator.

Question4 Labactivitiesa. Readthroughallofthelabstepsandidentifythestep(orsub-step)thatyouthinkwillbethe

mostchallenging.b. Listatleastonequestionyouhaveaboutthelabactivity.

INTRODUCINGDIODES

V_in

33Ω

D1

V_out

A

9

Figure7b:DiodeinacircuitStep0 DeterminetheV-Icharacteristicforasilicondiode.

a. BuildthecircuitinFigure7b,witha1N4002silicondiodeforD1.b. MeasurethecurrentIandthevoltageacrossthediode(NotethatthisisVout)fordifferentvalues

ofVinfrom−3Vto+3V.YoumayfinditisbesttouseaDMMtomeasureVoutandthepowersupplytomeasureI.Besuretotakemanydatapoints(atleast10)tomapoutthecharacteristiccurveofthediode,withafewextrapointsaroundVinof0.65-0.85V.Inthiscircuit,donotraiseVinabove3.5V!Thesmallresistorswehaveinthelabhavea0.25Wpowerrating,andathighervoltagestheywillgetveryhotandcanbedamagedorevenburnup.Howdoestheresultcomparewithyourpredictionsinthepre-labforanidealdiode?HowdoesitcomparewiththecartooninFig.1?

c. SpecificallyconsiderthepointsatVin=+0.5V,+0.8V,and+3.0V.Whatisthepowerdissipatedinthediodeandintheresistorforthesevalues?

d. Now,connecta1kHzsinewavewitha3Vamplitude(6Vpeak-to-peak)toVininplaceoftheDCvoltagesource.RecordtheresultingVoutwaveformandcomparewithyourprediction.

e. Optionalbutrecommended:Measurethewaveformofthevoltageacrosstheresistorinsteadofthediode.Thismightrequiresomeminorreconfigurationofthecircuit.DoesthevoltageacrosstheresistorlooksimplylikethedifferencebetweenVinandthevoltageacrossthediode?Ifnot,whatmightaccountforthedifference?

Step1 DeterminetheV-Icharacteristicforalight-emittingdiode.

a. Now,replacethesilicondiodewitharedLED(HLMPC625).RepeatStep0.b,butalsonotewhethertheLEDisemittinglightandgiveaqualitativedescriptionoftheamountoflight(none,verydim,moderate,bright,verybright).HowdoestheLEDbehavedifferentlyinthecircuitfromasilicondiode?DoestheLED’slightoutputseemtobeproportionaltovoltageorcurrent,orsomecombination?Anotherequipmentprotectionnote:readthespecsheetfortheLEDandbesurenottoexceedtheDCcurrentlimitfortheLED.

PHOTOMETER

Figure8:InvertingAmplifier Step2

Testtheamplifier(invertingconfiguration) a. Wewillfirstbuildtheamplifierinanormalvoltageamplifierconfigurationtotest

Rf

Vout

_

+

+15 V

-15 V

Cf

R

Vin

10

the setup and all the connections. Build the inverting amplifier circuit shown inFigure8.UseavalueforRF=RandclosetowhatyoufoundinLabprepproblem2, and a few pF of capacitance CF across the feedback resistor to avoidspontaneous oscillations. Draw the schematic in your lab book and record thevaluesofthecomponents.

b. Predict the low frequency gain and 3dB frequency of the circuit.HINT: Look atyourlapprepsolutions.ThiswillbetheRfCftimeconstantfrom(2bprelab);inthiscircuitthetimeconstantwillbedeterminedbythefeedbackresistorandcapacitor(not the input resistor).Design a test to make sure the amplifier is working aspredictedatlowfrequency(muchlowerthanthecutoff).Don’tbothertestingthehigher-frequencybehaviorthistime.

Step3 Roomlightphotometera. Reconfigureyourvoltageamplifiertothetrans-impedanceamplifiercircuitshown

inFig.6.Payattentiontothedirectionofthephotodiode.YoucanuseyourDMMsettothe“diodesetting”todeterminewhichsideofthediodesisthecathode.

b. Measuretheaverageintensityoflightfromthefluorescentlampsinthelabfromtheoutputofyourphotometercircuit.HowdoesyourresultcomparewithyourLabprepestimate?Keepinmindthattheestimateyoumadeofthelightintensitywasveryrough,andalsonotethatthedatasheetonlygivesa“typical”valueofthe sensitivity of the photodiode. How could you refine your model to moreaccurately represent your system or your physical system to more accuratelyrepresentyourmodel?List twopossiblerefinementsandcompleteat leastone.Explain why you think this refinement could allow you to getter bettermodel-measurement agreement. Report on what you did, your newmeasurements/model predictions, and if you were successful in getting betteragreement.

OPTICALCOMMUNICATIONLINK

Figure9:Opticalcommunicationlink.Left:LEDemittercircuit,Right:Photodetectorcircuit

11

Step4

LEDemitter a. Set up a light emitting diode typeHLMP-C625 as the transmitter on a separate

smallcircuitboardaccordingtotheschematic inFig.9.Youwilleventuallydrivethe inputvoltagewith the functiongenerator (See step3b.) Be sure toprotecttheLEDwithaseries resistance thatprevents the forwardcurrentexceeding20mA.HINT:YoucalculatedtherequiredresistanceintheLabprep.Also,connectaprotection diode in parallel with the LED but with opposite polarity. This willprevent you from accidentally running the LED at with a large negative biasvoltage,causingittobreakdown.

b. BeforeconnectingittotheLEDtransmittercircuit,setupthefunctiongeneratorto produced 1 kHz square waves with the upper voltage level at 10V and thebottomvoltageat0V.Youcanaccomplish thisbyusing theDCoffset settingofthe function generator. Now connect the function generator output to the LEDtransmittercircuit.CanyouseetheLEDturnon?

Step5

MeasuretheoutputoftheLEDwiththephotometer a. Place the LED transmitter at a distancewhere you expect ameasurable signal

fromthephotodiode.Useyourprelabcalculationtodetermineagooddistance.Whatcriteriadidyouuse todetermine thisdistance?The light fromtheLED isdirectedforwardsomakesuretoorientboththeLEDandphotodiodesotheyarepointingateachothertomaximizetheamountoflightdetected.Youcancheckalignment by using a piece of white paper to see if the red illumination iscenteredonyourphotodiodedetector.

c. Observetheinputdrivingsignalandtheoutputofthereceiveronthescopeusingdccouplingforbothsignals initially. Makesurethereceivedsignal isduetotheredLED,andnotfromanelectricalconnection.Howdidyoudeterminethatiswasworking? You may eventually want to switch to AC coupling on the scope toremovethelargeDCoffsetduetotheroomlights.

Step6 Optional(5pointsbonus):ComparewithPredictionandRefineModel a. Measuretheintensityofthetransmittedlightandcomparewithyourprediction

versusdistance.Howdoesthebackgroundlightfromtheroomaffectthesignal?Explain how youmade thismeasurement. Does yourmeasurement agreewithyourprediction?

b. Stateexplicitlywhat criteria youused todetermine if theyagree. If yourmodelpredictions andmeasurements do not agree, list all of themodel and physicalsystemrefinementsyoucouldtrytogetbetteragreement.Tryatleastoneoftherefinements. Explain why you think this refinement could allow you to getterbetter model-measurement agreement. Report on what you did, your newmeasurements/model predictions, and if you were successful in getting betteragreement.