Uncertainty Estimation of Spectral Mismatch Correction Factor for Incandescent Lamps

Transcription

1 ISSN: Volum 3 Numbr 7 (July-015) pp Uncrtainty Estimation of Spctral Mismatch Corrction Factor for Incandscnt Lamps Manal A. Haridy* hotomtry and Radiomtry Division, National Institut of Standards (NIS), Giza, Egypt *Corrsponding author KEYWORDS Spctral Mismatch corrction Factor, Lamp Sasoning, Uncrtainty, Total luminous flux, Commrcial Incandscnt Lamp, Intgrating Sphr hotomtr A B S T R A C T In th prsnt rsarch, uncrtainty of spctral mismatch corrction factor of total luminous flux masurmnts ar stimatd for som typs of commrcial incandscnt lamps. A st of ight incandscnt lamps hav bn sasond for 60 hours and th aging curvs hav bn drawn for ach lamp. A st up basd on NIS Spctroradiomtr ocan optics HR 000 with uncrtainty 4.7% and th photomtric bnch has bn usd for masuring th spctral powr distribution of th lamps to dtrmin th spctral mismatch corrction factor and stimat th uncrtainty du to this factor. Anothr st up basd on.5mtr NIS intgrating sphr and NIS photomtr LMT U1000 with uncrtainty 4% and a group of NIS luminous flux standard lamps calibratd at National hysical Laboratory in England (NL) with uncrtainty 0.8% ar usd to masur th total luminous flux for th ight incandscnt lamps during th lamp sasoning. This rsarch dmonstrats and ralizs th procss rquird to dtrmin th spctral mismatch corrction factor and to stimat th uncrtainty du to it for som commrcial lamps to improv thir total luminous flux masurmnts. Th spctral mismatch corrction factors ar varid from approximatly 0.13% whn using th 400 K standard lamps to approximatly 0.14% whn using th 750 K standard lamps. Also, th uncrtainty stimation for spctral mismatch corrction factors ar varid from to dpnding on th corrlatd color tmpratur of th standard lamps. Introduction Th incandscnt lamp is th oldst of th major lctric light sourc familis. Th viability of this sourc ovr th yars is a consqunc of its many advantags and rlativly fw disadvantags. Simplicity, low cost construction, and xcllnt color rndition ar major attributs of this class of light sourc [1].Th incandscnt lamp is a lamp that crats radiant nrgy whn its mtallic filamnt is hatd by an lctric currnt. Th filamnt is dsignd to produc radiant nrgy in th visibl portion of th 6

2 lctromagntic spctrum (light). Th filamnt is of a spcial matrial that is supportd in an nvlop (bulb) that has bn vacuatd or filld with an inrt gas such as argon, nitrogn, or krypton. In addition to light, th hatd filamnt mits infrard and ultraviolt nrgy. Whn ithr of ths radiations is accntuatd, th lamp may b usd as a sourc of that nrgy. Th important parts of an incandscnt lamp ar th bulb (nvlop), th filamnt, and th bas. Th bulb may b clar, colord, insidfrostd, or coatd with diffusing or rflcting matrial. Most lamps hav softglass bulbs; hard glass is usd whn th lamp will b subjctd to suddn and svr tmpratur changs. Lamps hav a varity of bulb shaps, bas typs, and filamnt structurs. Luminous flux lamps should b handld carfully to avoid mchanical shocks to th filamnt. Thy ar normally opratd in th bas-up position and sasoning should b conductd in th sam burning position. Th lamp must always b opratd in th sam burning position. Bfor luminous flux working standards lamps ar submittd for calibration, thy must b sasond and tstd for stability. Ths incandscnt lamps ar usually opratd on DC powr in ordr to hav much bttr stability of powr supplis, bttr accuracy of lctrical masurmnts than on AC powr and to avoid problms of flickr undr AC opration. During th sasoning procss, th tungstn filamnt of th lamp is partially rcrystallizd according to th lctrical polarity applid. If th polarity is changd, th lamp filamnt may undrgo a chang of stat. Thrfor, th polarity should b kpt th sam whn th lamps ar sasond and whn thy ar usd. Th polarity should b markd on th lamp bas or it should b clarly dfind in a documnt []. Th lamps intndd to b usd for masuring photomtric, colorimtric, and lctrical charactristics should b sasond [3] until any tim-rlatd chang in ths charactristics bcoms small. For nw lamps of almost vry typ, th gratst rats of chang in th oprating charactristics gnrally occur during th first hours of opration. Normal sasoning is gnrally prformd at ratd voltag for a minimum priod of onhalf of 1 prcnt of ratd lif. Lamps should b sasond in th bas-up position. Th position usd during sasoning must b maintaind throughout subsqunt tsting and it must b notd in th tst rport. Acclratd sasoning is not rcommndd for tungstn halogn lamps, and is not rcommndd for lamps having a ratd lif of 100 hours or lss. Th following two quations [4,5] allows th dirct calculation of acclratd sasoning minuts (ASM): (0.005 Ratd Lif) ASM 60 (1) 13 ( rcnt Volts) (0.005 Ratd Lif) ASM 60 () 4 ( rcnt Amp) whr : Th numrator ( Ratd lif) is th minimum normal sasoning hours for th lamp that is to undrgo acclratd sasoning. rcnt volt and prcnt ampr ar xprssd as th dcimal, voltag or currnt ratio. For xampl, if a voltag ratd lamp is oprating at 10 prcnt of ratd volts, th dominator in Equation (1) should b valuatd as Equation () [5]. In photomtry, th comparison dvic is th photomtr. Sinc no photomtr prfctly 63

3 matchs th V ( ) curv, w nd to dtrmin how closly w approach th V ( ) curv. Th output of a photomtr is a singl numbr; th intgration ovr th spctral rang is all don intrnally and on numbr coms out. Most photomtrs ar calibratd using what is calld th CIE Illuminant A, which is spctrally smooth light sourc, similar to a blackbody radiator oprating at 856K [6]. Whn th photomtrs masur light sourcs whos spctral distribution is diffrnt from th CIE Illuminant A, an rror occurs du to th spctral mismatch of th photomtrs. This rror is corrctd by a spctral mismatch corrction factor [7]. An rror occurs whn a photomtr masurs a light sourc having a spctral powr distribution diffrnt from th standard sourc for which th photomtr was calibratd. In ordr to corrct for such spctral mismatch rror, th photomtr must b charactrizd for its rlativ spctral rsponsivity. Th spctral mismatch corrction factor SCF is givn by [8]. V ( ) : is th spctral luminous fficincy function, which dfins a photomtric masurmnt. Uncrtainty of th dtrmination of th spctral mismatch corrction factor u(scf) which can b dtrmind rgarding to Equation (3) and according to rfrnc [9, 10] by th following quation: u var iabl ( Hnc w hav u ( SCF) Whr (5) is SCF ) variabl ( is u it (variabl) ( it (4) Ri ( R is : th summation of S ( ) within th visibl wavlngths rang. T it : th summation of ( ) within th visibl wavlngths rang. R i : th summation of R ( ) within th visibl wavlngths rang. i 830nm T ( ) V( ) d 360 all wavlngth s SCF 830nm T ( ) R( ) d all wavlngth s 360 (3) whr S S ( ( ) ) R( V( ) d ) d Thn from quation () w hav u( SCF) ( is (6) is ( it it R( i R i T ( ) : is th rlativ spctral output of th tst sourc. S ( ) : is th rlativ spctral output of th standard sourc. R ( ) : is th rlativ spctral rsponsivity of th photomtr. Mthods and xprimntal st-up Aging lamps All th total luminous flux lamps ar clar round bulb. All lamps ar from NIAZA Company in Egypt, thy ar thr groups of diffrnt watts. Th ight lamps wr ach st to oprat at a corrlatd color 64

4 tmpratur (CCT) for ach lamp as shown in Tabl (1). Th lamps sasond or agd for approximatly 60 hours to b sur from th photomtric charactrization stability for ach lamp. Th lamps wr agd by using th NIS stup of aging lamps as shown in Figur (1). Th data was masurd vry tn hours. Masurmnt St-up of th total luminous flux quantity during sasoning th lamps Th NIS photomtr usd is LMT U1000 and th intgrating sphr [11] usd has a diamtr of.5 mtr and is coatd intrnally with a uniform layr of barium sulfat ( BaSO ).Th spctral quality factor 4 ' f for th photomtr has bn dtrmind 1 for th sphr-photomtr with valu 6.5% [1]. Th flux lamps ar mountd in a basup configuration at th cntr of this sphr into a lamp sockt supportd from th top of th sphr. Figur 1, shows th.5 mtr diamtr intgrating sphr Masurmnt St up of th spctral powr distribution of th lamps Th St up of masuring th spctral powr distribution lamps is in Figur. 3. It masurd dirctly using th photomtric bnch and th Spctroradiomtr ocan optics HR 000 at NIS with uncrtainty 4.7% [13,14]. Rsults and Discussion Th rsults of th lamp aging ar prsntd in Figurs (4) to (11). Th quation of th fittd lin in Figurs (4) to Figur (11) rprsnt th aging rat of lamps. Th quation in ach figur contains two valus and t, whr is th normalizd photomtr rading and t is th aging tim for this rading. is a function of th slop (th rat of changing of lamp output pr unit tim) and th intrcpt ( valu at t=0).ths curvs ar vry usful to dtrmin th uncrtainty du to th aging of th lamps whn w us thm as working standard lamps. Th lctrical control paramtrs of th OSRAM Luminous Flux Standard Lamps and NIAZA lamps for th masurmnts prformd ar givn in th Tabl 1 and Tabl. Uncrtainty Modrn mtrology practic in any fild rquirs accurat stimation of uncrtaintis. Th concpt of uncrtainty as a quantifiabl attribut is rlativly nw in th history of masurmnt, although rror and rror analysis hav long bn a part of th practic of masurmnt scinc or mtrology. It is now widly rcognizd that, whn all of th known or suspctd componnts of rror hav bn valuatd and th appropriat corrctions hav bn applid, thr still rmains an uncrtainty about th corrctions of th statd rsult, that is, a doubt about how wll th rsult of th masurmnt rprsnts th valu of th quantity bing masurd. Consistnt stimation of uncrtainty according to th mthods dscribd in th ISO Guid to th Exprssion of uncrtainty in Masurmnts [15] is a major concrn of modrn mtrology. This applis not only to th dirct masurmnt of Intrnational Systm of Units quantitis, but also to masurmnts drivd from thm. All uncrtaintis ar givn as fractional uncrtaintis and as on standard dviation (covrag factor k=1). Th rlativ xpandd uncrtainty (k=) was obtaind by assuring a normal distribution and multiplying th rlativ combind standard uncrtainty by a covrag factor of k=. 65

5 Tabl.1 Th lctrical control rsults of NIS standard lamps SET Color Total luminous NIS Standard Voltag Currnt tmpratur flux Lamps (amprs) (Volts) (Klvin) (lumn) NIS-E NIS-E NIS-E NIS-E NIS-E Tabl. th lctrical control rsults of commrcial incandscnt lamps NIS SET Color Total Voltag Standard Currnt tmpratur luminous flux Lamps (amprs) (Volts) (Klvin) (lumn) NIS NIS NIS NIS NIS NIS NIS NIS Tabl.3 Spctral Mismatch Corrction Factors (SCF) for standard lamps Lamp Spctral Mismatch Corrctions NIS-E NIS-E NIS-E NIS-E NIS-E NIS-E Tabl.4 Spctral Mismatch Corrction Factors (SCF) for Tst Lamps. Lamp Spctral Mismatch Corrction NIS NIS NIS NIS NIS NIS NIS NIS

6 Tabl.5 Th uncrtainty valus of th total luminous flux du to th spctral mismatch corrction factor Lamp NIS-E1 NIS-E NIS-E4 NIS-E31 NIS-E3 NIS-E33 NIS NIS NIS NIS NIS NIS NIS NIS Figur.1 Th NIS Aging systm for Lamps photomtr at NIS, Egypt Figur. Th NIS intgrating sphr photomtr systm st up for th luminous flux masurmnt 67

7 Figur.3 St up of masuring th spctral powr distribution of NIS total luminous flux working standard lamps Figur.4 Th aging curv of NIS

8 Figur.5 Th Aging Curv of NIS--00 Figur.6 Th aging curv of NIS-3-00 Figur.7 Th aging curv of NIS

9 Figur.8 Th aging curv of NIS--150 Figur.9 Th aging curv of NIS Figur.10 Th aging curv of NIS

10 Figur.11 Th aging curv of NIS--100 NIS photomtr LMT U1000 with uncrtainty 4%, NIS Spctroradiomtr with uncrtainty 4.7% and NIS-standard lamps calibratd at National hysical Laboratory in England (NL) with uncrtainty 0.8%. Uncrtainty of sphr calibration photomtr factor (0.15%) and hotomtr Spctral ' Quality Factor f 1 qual 6.5% [1]. Th spctral mismatch corrction factors ar varid from approximatly 0.13% whn using th 400 K standard lamps to approximatly 0.14% whn using th 750 K standard lamps. Also, th uncrtainty stimation for spctral mismatch corrction factors ar varid from to dpnding on th corrlatd color tmpratur of th standard lamps as shown in Tabl 4 and Tabl. Conclusions Th aging curvs prsntd in Figurs (4) to (11) ar vry usful to dtrmin th uncrtainty du to th aging of th lamps whn it usd as working standard lamps. Th aging of th lamps to dtrmin whthr thy wr adquatly stabl for us as standard lamps. It was found that th typical aging tim of 48 hours was sufficint for th incandscnt lamps usd, and that by rcording th chang in lamp output with tim w could dtrmin that th lamps had stabilizd to a uniform aging rat, and that w could us this data to stimat an uncrtainty in th calibration of th lamp du to th aging of th lamps. Th aging data could also b usd to dtrmin whn th lamp should b rcalibratd in ordr to maintain a dsird accuracy for th total luminous flux output of th lamp. Th spctral mismatch corrction factors ar varid from approximatly 0.13% whn using th 400 K standard lamps to approximatly 0.14% whn using th 750 K standard lamps. Th uncrtainty stimation for spctral mismatch corrction factors ar varid from to dpnding on th corrlatd color tmpratur of th standard lamps. Rfrncs 1.CIE 77, Elctric Light sourcs- Stat of Th art (Commission Intrnational d l Eclairag, 1987, Vinna, Austria..Yoshi Ohno, 1997, Handbook of Applid hotomtry, Masurmnt rocdurs, Amrican Institut of hysics, Woodbury, Nw York. 71

11 3.IESNA (Illuminating Enginring Socity of North Amrica), Guid to lamp sasoning, LM-54-99, IES, Rsarch papr No. 50, National Burau of standards. 5.IESNA (Illuminating Enginring Socity of North Amrica) Lighting Handbook, 8 th Edition, Light sourc sction, Nw York: Illuminating Enginring Socity of North Amrica, A.A. Gartnr, Basics of hotomtric Calibration, hotomtry, Radiomtry And Colorimtry Cours NRC, Ottawa, Canada, Ohno, Y., hotomtry Calibration, NIST Spcial ublication 50-37, A.A. Gartnr, LED Masurmnt Issus, hotomtry, Radiomtry And Colorimtry Cours NRC, Ottawa, Canada, Guid to th Exprssion of Uncrtainty in Masurmnt, First Edition, Intrnational Organization for Standardization (ISO), Manal. A. Haridy, Th Ralization of Luminous Flux Scal and Its Application in rparing Calibratd and Tstd Lamps, Doctoral thsis, collg of Womn, Ein Shams Univrsity, Egypt and NRC Canada, A.B. El-Bialy, M.M. El-Ganainy and E.M. El-Moghazy, A comparativ conomical study of roadway lighting systms using High Intnsity Discharg (HID) lamps in Egypt J. Basic. Appl. Sci. Rs., (4) , Manal A. Haridy, 015, Improvmnt uncrtainty of total luminous flux masurmnts by dtrmining som corrction factors, Int. J. Curr. Rs. Aca. Rv; 3(6), pp A. l Bially, M.M. El-Ganainy and A. Elkaml, Color Rndring Uncrtainty Australian Journal of Basic and Applid Scincs, ISSN , 4(10): , A.B. El-Bialy, M.M. El-Ganainy and E.M. El-Moghazy, Uncrtainty dtrmination of corrlatd color tmpratur for high intnsity discharg lamps Natur and Scinc., 9(3), Guid to th Exprssion of Uncrtainty in Masurmnt, First Edition, Intrnational Organization for Standardization (ISO),