View AB20-6_4641205.PDF datasheet online --- IC-ON-LINE

Datasheet File OCR Text:

application brief ab206 replaces an 114 9 6 reliability considerations for superflux leds leds provi d e solidstate li g h ting a n d a r e ther efor e extr emely d u r a bl e . if the recomm ended soldering and ope rati ng c o nditi o ns a r e f o llo wed, su pe rfl u x leds will survive for the life of the vehi cle. superflux le ds have perfo r med wel l in a numb e r of d u rab ility, reli abili ty, and ac cel er ated lif e tests. some of these tests are sum m a r iz ed in l u mil e d s super f lux led reliability data sheet. a current revisio n can b e obtai n ed by contacting your lo cal fiel d sal e s engineer or at the fo llo win g url: http://w w w .lum ile ds.c om this secti o n is dedi cate d to comm uni cati n g: 1) the major tests included in the reliability data sheet, 2) the app r o p ri ate testin g of assembli es co ntaining sup e rflux leds, and 3) p r ovid i n g info rm atio n regardi n g the ty pical ch an ge in su pe rfl u x led light o utput over typical appli catio n lifeti m es. please no te that the reliabilit y data sheet also incl ud es the c u mul a tive sa mple size and failure ra te (if other than zero). table of contents durability, reliability, & accelerated life tests 2 mechanical shock 2 v i b r a t i o n 2 operating life 3 weathera b i l i t y 3 corrosion resistanc e 3 temperature cycle 3 simultaneo us power and temperature cycle 4 solder heat resista n ce strife tes t 5 recommended led assembly va lidation tests 6 verifying root causes o f power & temperature c y cle failures 6 changes in light- outp ut during the operating life of alingap superflux leds 8 1
table 6.1 reliability , du rability a n d ac cele r a ted life te st s m echanical s h o c k milstd8 83 me tho d 2002 co ndi t ion b: 3 shocks e ach xy z a x is, 3 000g, 0.3 ms vib r ation variabl e frequency milstd8 83 me tho d 2007 co ndi t ion a: 4 cycles, 4 minutes, eac h x yz axis, 20g minimum, 20 to 2 000 hz vib r a t io n sae 575 dec88 section 4. 1: 10 to 55 hz, 1 mm peak t o pea k , 2 mi nutes per cy cl e, 60 mi nu tes total dura ti on oper ating life tests: con d it ion a: t a = 100 c con d it ion b: t a = 85 c con d it ion c: t a = 55 c con d it ion d: t a = 25 c con d it ion e: t a = 40 c all tests of durati on 1,000 ho urs i f = 20 ma (as)/3 0 ma (ts); t j ~ 11 0 c i f = 45 ma (as)/5 0 ma (ts); t j ~ 11 0 c i f = 70 ma; t j ~ 90 c i f = 70 ma; t j ~ 60 c i f = 70 ma; t j ~ 5 c weather a bility con d it ion a: for w ard con d it ion b: rev e rse 85 c, 85% rh, 4 5 ma, 1000 h 85 c, 85% rh, 5 v, 1000 h sa lt at mo s p he re milstd8 83 me tho d 1009: 35 c, 48 h co rro s i o n sae 575 dec88: 25 c , 2 ti mes 24 h w i th o n e ho ur dry tem p er atur e cycle conditio n a: 55 to 100 c conditio n b: 40 to 120 c conditio n c: 40 to 100 c milstd8 83 me tho d 1010: 15minute dwell, 5minute trans f er, 300 cycles 15minute dwell, 5minute trans f er, 300 cycles 15minute dwell, < 10 s tra n sfer, 3 00 cycles p o wer & tem p eratur e cycle conditio n a: 70 ma con d it ion b: 50 ma 40 c t o 85 c, 3 c per minute tra n sfer rate, 2 hour cycle 70 ma, 5minute on/o ff, ~ 22 5 c/w therm a l resista n c e (t j ~ 120 c), 50 cycles 50 ma, 5minute on/o ff, ~ 22 5 c/w therm a l resista n c e (t j ~ 110 c), 50 cycles s o lder heat r e sistance s t r i fe refere nce figure 6.3 durability, reliability, & accelerated life tests mechanical shock lumi l eds su p e r f lux leds a r e e n c a psu l at ed in a solid pl astic p a c k a g e. th er e f ore, they a r e extrem ely ru g g ed a n d e a sil y survive eve n extrem e me c h a n ic al sho c k tests. as shown in tabl e 6.1, lumileds us ed the test conditi o ns specifi ed i n milstd 883 metho d 200 2 co nditio n b to valid ate the superflux le ds mec h a n ic al s h oc k p e rfo r m a n ce. this tes t was conducted three times in each di rection at 3000 g for 0.3 millis econds. all of the superfl u x leds subjected to this test survived without damage. vibration superflux le ds perform ex tremely well i n typical vib r atio n tests (u p to 2 k h z). superflux leds have survived mi lst d883 m e tho d 2007 co nditio n a and sae 577 vi brati o n tests. ultrasonic w e l d ing is freq u e ntly used to sec u r e the lamp housing to the lens and is occasi o nally used to secure the pr inted circuit assembly cont aini ng th e leds to mounting p o sts in the lamp housi n g. althoug h ultrasoni c w e ldi n g of 2
the lam p ho using to the le ns has not pr od uc ed any kno w n le d failur e s it ha s been i d e n tified as the ro ot caus e of some catastrop h ic failures of resistors and other le aded components. in additi on, ultrasonic wel d ing of mounti n g p o sts has be en i d e n tified as the r o ot caus e of some catastr oph i c f a ilu res of led s . in these ca ses, the transfer of ultras o n ic en ergy fr om the mounting posts resulted in a bro k en wi re withi n the imme diat e l y adja ce nt le ds. although no failures of sup e rflux leds h a ve be en attri b uted to ultrasonic welding, the p o tential for thi s failure mode still exis ts. for this reason, heatst a ki ng o r a low e rfrequency join ing process such as vibrati o n w e ldi n g is pr efer re d for sec u ri ng the mounting posts of the la mp housing to the pc b. in cas es wh er e ultr asoni c w e ldi n g must b e used, it is extremely impo rtant to build many evaluation units to prov e the desig n ?s ability to withs t and ultras onic vibrations. operating life beca use su p e r f lux leds a r e soli dstate devices, they survive th ou sa n d s of hour s of operatio n . as summarized in tab l e 6.1, lumi l eds us e d sever a l am b i ent temp er atur es to valid ate the o p er atin g life p e rfo r ma nc e of the superflux le ds. these le ds perform w e ll over their o p e r ati n g tempe r atu r e ran g e of ?40 weatherability superflux le ds also perfo r m well i n weatherability tests. as summariz e d in t a ble 6.1, lumil eds used b o th for w a r d bias ed and reversebiased humidity tests to validate perf orm a n ce after exp o su r e to humi d ity. thes e tests were performed with th e l e d s m o un t e d on a pc b. n o housing was used for this test. corrosion resistance the m e tal leadfram e used i n sup e rflux leds is mad e from a cop p e r alloy. the p o rtio n of the le adfr ame th at is in side the plastic body is first plated with nickel and th en silver. the po rtion of the le adfr ame that is outside the pl astic b ody is coat ed in tinl ead sol d e r . the r efo r e, sup e r f lux leds perf orm well in corr osion resistance tests. the tw o tests listed in t a bl e 6.1 were used to valid ate sup e r f lux led p e rf orm a n ce d u ri ng corrosi o n testi n g. the mils t d88 3 test was perf orm e d usi n g individ u al leds, while the sae 576 test was perf orm e d usi n g leds mo unted o n a pri n ted c i r c uit boa r d. no housi n g w a s used fo r either test. temperature cycle the se mic o nd ucto r that emi t s the light in supe rf lux le ds, the leadfr ame th at it is mounted on, the bond wi re that connects the ano d e of the chi p to the an ode of the lea d fra m e, an d the plasti c that forms the led?s body all have different co effi cients of thermal expa nsio n. th er efor e, therm a l cycl e test perf orm a n ce i s an impo rtan t measur e of 3
superflux le d durability. o n e of the two typical w e a r ou t mech an isms in th erm a l cyc l e te sting is sepa ratio n be twee n the pl a s tic enc a ps ul ant an d the sides of th e refl ecto r cu p in the le adf r ame as show n in fi gure 6.1. this type of sep a ration lifts the semi con d ucto r die off o f the leadf r am e an d ca uses an o p e n ci rc uit or i n termittent o p e n cir c uit. th e se con d typica l failure m e chanism is a ne cked do w n wi re b r e a k above th e bal l bo n d (wire bond) as shown i n fig u re 6.2. in ord e r to avoid th ese fa ilur e me ch ani s ms, it is importa n t that the maxi mum ju nctio n tempe r atu r e li sted in the sup e r f lux led tec hni cal data sheet is not exce ed ed (r efer to appli cati o n bri e f ab20 4 therm a l management consideratio ns for superflux leds . lumileds used th e thre e tempe r atu r e c cle tests listed in ta le .1 to alid ate sup e r f lux led p e rf orm a n ce afte r therm a l c cle testing. figure 6.1 exa m ple of led di e delami natio n . figure 6.2 exa m ple of nec k - d own wire brea k. simultaneous power & temperature cycle simu ltan eou s ly c y c ling both pow er an d tempe r atu r e c a n a c c e le rat e typical l e d fai l ur e modes. altho u gh th ese c o nditio ns ar e m u c h mor e extrem e than those se en in actu a l fi eld use, this type of test is very hel p ful in com p aring the pe rfor ma n ce of different designs. perfo r ma nc e und er th ese c o nditi o ns is benchmarked duri ng initial led pro d uct va lida t ion an d th e n ch ec ke d du r i n g su bsequ e n t valid ation test s for pro cess or mate ri al ch ang e s. this test should also be used when validating assembl i es co ntain i ng s u pe rfl u x leds to ensure that excessive se lf heating will not result in pr emat ur e failu re of su pe rfl u x leds. although sae j188 9 des c ri bes a po wer tempe r atu r e c y cle test, lum ileds us ed mo re extrem e po w e r temp er atu r e cycl e co nditi o ns to valid ate sup e r f lux led p e rf orm a n ce. sae j18 89 recomm ends pow eri n g the device w h ile wa rmin g and hot an d leavi n g it unlit wh en cool ing and cold. lum ileds? tests, which are summarized in tab l e 6.1, are simil a r to s a e j18 89, but th e pow er is cy cle d by exces sive solde r he at, or impr op erly man u fa ctur ed supe rf lux le ds. lumileds also extends the test from the 2 5 c y cles r eco mm end e d i n sae j18 89 to a minim u m of 50 cycles. 4
5 solder heat resi stance s t rife test supe rf lux le ds have be en design ed so that heat generated within the di ode is quickly dissipated out through the copp er leads. this therm a lly effi ci ent desi gn al l o ws these le ds to be driven at hi gher po wer levels relative to other leds, thus re sulting i n opti mal opti cal perf orm a n ce. ho weve r, wh en the l e a d s a r e immersed in solder, this des i gn als o allo w s heat to conduct from the le ads up into the active ar ea w h e r e th e diod e is loc a ted. bec a us e supe rf lux le ds transfer h e at from sold er todi e much more quick l y than other leds, it is extrem ely imp o rta n t that rec o mme nd ed solde r profil es are fol l ow ed duri ng assembly to avoid thermal stress ing of the leds. lumileds recomm ended soldering pr ofiles are lo cated in appli catio n note 114 92 me ch an ica l de sign co nside r ati o n s for supe rfl u x leds . lumil eds uses the sold er h eat r esist a n ce strife tes t shown in figure 6.3 to verify that supe rf lux led s can be soldered without da mag e . the tem p e r atur e cycl es and ho t function al te st ensur e det ectio n of latent d a mag e or int e rmi ttent connections. this test sequence was used to valid ate sup e rflux led perform a nce. in ad dition, this test is rep eat ed r e gu larly to mo nito r the pe rfo r ma nc e of superfl u x leds. table 6.2 r ecom m e nded test test pa ra me te rs p o sts older i ng 30x v i sual inspection refer to application note 114 92 mechanical design consid e r ations for superflux leds for a description of the procedure and accept/reject criteria. r oom tem p er atur e oper ating life ta = 25 c, 5 0 0 ho urs at no minal driv e c o ndit io ns tem p er atur e cycle 40 c to 85 c [1] , dwell & transfer times selected based on therm a l mass of th e assembly, 50 cycles p o wer tem p er atur e cycle 40 c to 85 c [1] , 3 c per minute transfer rate , 2hour cycle, 5minute on/off, 50 cycles at no minal driv e c o ndit io ns. or, sae j18 89 o c t9 3: 40 c to 85 c [1] , 0.6 c to 5 c per min u t e transfer rate, onehour minim u m dwell, powered at no minal driv e conditions whil e warming and hot, no power while cooling and col d . note: the s e co ndi t ion s e x cee d l u m iled s r ecom m en ded solde r in g profile s. figure 6.3 sol d er heat resista n ce strife te st. table 6.2 reco mmende d vali d a tion tes t ing .
recommended led assembly validation tests lumi l eds r e c o mmen d s that custom ers us e the tests listed in the table 6.2 to validate the dur a bi lity of assembli es co ntain i ng s u pe rfl u x leds. other t e sts (such as vibrati o n, or cor r osi o n r e si stanc e ) may b e inc l ud ed to ch ec k the pe rfor ma n ce of other co mponents, materials, or in terconnections in the assembly. pleas e re alize that lumi l eds ca nnot g u a r a n tee led perform a nce during durab ility tests if the maximu m op er ating o r stor age tem p e r at ur e ranges, or the maximum junction temperature listed in the s u perflu x l e d te ch nica l da ta she e t are exc eed ed. in so me ap plic atio ns, such a s h i gh m ou nt sto p lamps mounted in the he adli ne r be hi nd r e a r win d o w glass, the l o ca l ambi ent temp er atur e ca n r e ac h tempe r at ur es in excess of 10 0 c. lu mile ds r eco mme nds that lig h ting syste m suppli ers w o r k with ve hicl e man u fa ctur er s to perform t e mp er atur e meas ur eme n ts on simila r ve hicl es or mo c k u ps to determ in e the a c tua l ope r atin g an d storag e conditio ns to whi c h the le d assembly w ill be expose d. duri ng thes e stud i e s it is important for the lighting system suppl i e rs to rec o rd b o th tempe r atu r e a n d sup p ly voltage m eas ur e m ents. both of these parameters wi ll be required when desig n in g ass e mbli es for w orst case c o n d itions (reference ap plication brief ab203 el ect r i cal design consi d er atio ns for supe rf lux le ds. ) verifyin g root causes of power & tempe r ature cycle failures the m o st co mmon c a us es for supe rf lux led failu res d u ri ng powe r & tem p er atu r e cy cl e testing ar e ex cessive h eat dur i ng sol d e r i n g, therm a l sho c k, and exc e e d i n g the d e vic e s absolute maximum junc tion temperature. sepa rati on b e twee n the le a d fram e a n d th e en caps u la nt is usually c a us ed by exc e ssi ve solder heat or thermal sh ock . if this type of failu re is susp ecte d , the roo t cause c a n b e confi r m e d by meas uri n g the wave sold er station tempe r atu r es and th e temp er atur e gr adi e nts that the superflux le ds were exposed to during assem b ly. this is accomp lished b y attach ing a therm o co upl e to the le ad of an led on the sol d e r s ide of the pc b and r eco rdi n g its te mper a t ur e pr ofile dur i ng ea ch oper a t ion. th is shoul d in clu d e any p r eli m in ary op er ations such as surfa ce mo unt glu e c u r e , the solde r in g ope r atio n, any subseq ue nt rewo r k or tou c hu p oper a t ion s , an d an y po stsolder heating or cool ing cycl es such as c o nf orm a l co ating cu re. (reference ap plic ation note 1149 2 mec h ani cal design consi d er atio ns for supe rf lux le ds for pr ec au tion s re quire d wh en attach ing th er mo cou p les to the leads of led s .) the he atin g and cool ing rates must not exce ed 3 c pe r seco nd, and th e pr eh e a t and sol d e r i n g temp er atur es shoul d follo w the recom m endati ons provi d ed i n appli catio n note 114 92 me ch an ica l de sign co nside r ati o n s for supe rfl u x leds. exceeding the maximum ju nction temperature of the sup e rfl u x leds during power & temp er atu r e cycl e testing usua lly res u lts in a ne cked d o w n wire b r e a k ab ove the b a ll b o nd o n 6
the led chip ( s ee fi gure 6.2 ) . to verify this root ca use, the ju n c tion tem p e r a t ure d u ri ng a t y pica l powe r & tem p er atu r e cy cl e shou ld be meas ur ed. the m a ximu m juncti on temp er atur e ca n b e dete rmin ed b y addin g the temp er atur e differ e n ce b e twe e n the led lead a n d th e led junction to the maximum le ad te mper a t ur e. th e tempe r atu r e d i ffere nc e bet w ee n the led l ead and the led junction is the product of the forw ar d cu rr e n t flowin g through the led, the forw ar d voltag e of the led (meas ur ed at this forw ar d cu rr e n t), and the th erm a l r e sistan ce of the sup e rflux led (listed in the technical data sheet). this is express e d as the following equation: t j = t p + r = t p + ( r ? t at est ) wher e: t am ax = maxim u m ambi ent te mpe r atu r e fr o m powe r & temp er atur e cycl e test t ates t = ambie n t temper atu r e of ben c h test 7
changes in lightoutput during the operating life of alingap superflux leds the li ght outp ut of superfl u x leds will gradually change d u ring their lif etime. thi s chang e is usually a grad ual red u cti o n in li ght output; however it is al so possible to have a slight increase in light o utp u t during early operatio n follo wed by a g r adual reduction. this change in lightoutput is call ed lig hto utput deg r adatio n. typical am ounts of lighto utput d e gr ad a t i o n fo r so me po pul a r au to mo ti ve appl ic ations a r e summ ar ize d in ta ble 6.3 . light output d e grad ation o c curs most quickly during the initi a l ho urs of operati on and then slows with time as show n in fi gure 6.4 . there is some variation within a batch an d be tw ee n ba tc he s of superflux le ds. figure 6.5 shows that the amo unt of ligh t output d e gr adati o n is pro p orti onal to forw ard cu rrent. this figure also shows that some batches o f superfl u x leds get slightly bri g ht er b e for e start i ng the typi cal dec lin e. ambi ent temp er atu r e, hu midity, a n d sunli g ht hav e minim a l effe ct on light o utp u t de gr ada t ion as sh ow n in table 6.4, table 6.5, and fig u r e 6.6. for su pe rfl u x leds, lum ileds is ca refu l to sele ct aling a p materi al wit h an aver ag e lig ht o utput d e gr ad ation of less tha n 35% after 1 0 0 0 hours of op eration at 70 ma. table 6.3 lif e time cha n ge i n light- o u tp ut fo r su per f lu x le ds in a u tomotive si gnali ng a pplications lifetime lightoutput change @ t a = 25 c; i f = 20 ma (t ail) and 60 ma (stop and turn) f u n c t i o n o p e r a t i n g l i f e (h ) typical lop cha n ge tail 1,500 t o 3,00 0 5 to 5 % stop 750 t o 20 00 30 t o 5% tu rn 200 t o 70 0 20 t o 5% figure 6.4 op erating life test r e sults for hpwa- m hoo l e ds drive n at 7 0 ma, 55 c. figure 6.5 change in light- o utput over time for hpwt- dhoo driven a t mul t iple c u rren t s at 55 c. 8
lig h t output d e gr ad ation fo r supe rfl u x le ds is a function of the curr ent density in the led semicond ucto r chi p . the rat e of chang e w ill be low e r w h en superflux led s are d r iven at lower cu rr ents. figu re 6.5 sho w s how th e lig ht o utput varies over time for le ds driven at several currents. although this particul ar lot of le ds becam e slight ly brig hter with time at the lower forw ar d cu rr e n ts, their lig ht o utput p e a k e d and then b e g a n th e typical l oga ri thmic d ecli n e. at hig h currents superflu x le d lighto utput will pea k wit h in a few ho urs. at lower currents they may continue to get brig hter for sever a l h u n dr ed hours befo re begi nning the normal g r adual reduction over time. ambie n t temp er atur e has n e gli g ibl e effe ct on the lig htoutp u t deg r ad atio n of supe rf lu x leds. du r i n g th e in i t ia l p r o d u ct qualificatio n, hp monitored the change in lig htoutp u t du ri ng the ope rati ng a n d forwa r d bi ased w eath e r a bi lity tests listed in tabl e 6.1. th e ave r ag e c h a n ges for on e of the ts aling a p w a fers us ed for these tests are sho w n in t a bl es 6.4 and 6.5. t h e differ e nt stres s temper atur e s and the p r e s enc e of humi d ity ha d neg ligi b le ef fect on lig hto utput degradation. in fact, the 100
10 table 6.5 weathera bility te st s weather a bility tests a v er ag e c h ange in lightoutput (after 1.00 0 hour s) co ndit io n a , w h t o l, 8 5
11 company information lumi l eds is a wor l d class s u ppl ie r of li g h t emitting diodes (leds) producing billio ns of led s annually. lumileds is a full y integrated s u ppl ier, prod ucing cor e led mat e ri al in a ll thr e e base colo rs (red, gre e n, blue) and white. lumileds has r&d deve lopment centers in san jose, cal i fornia and best, the netherl a nd s. pro d ucti on cap a bilities in san jose, california and malaysia. lumi l eds is pi one e r i ng th e hig h flux led technol ogy and brid ging the gap betw ee n solid state led techno logy a n d the lig hting wo rl d. lumileds is absolutely dedicated to br i n gi ng the b e s t and bri g htes t led technology to en abl e ne w a ppli catio ns an d mar k ets in t h e li ghti ng w o rld. lumileds www.luxeon.com www.lu mileds.c om for technical a ssista n ce or the locatio n of your nearest lumile ds sales office, cal l: wo rld w ide: +1 408- 435- 6044 us toll free: 87 7- 298- 9455 europe: +31 49 9 339 439 a sia: +65 6248 4759 fax: 408- 435- 6 855 email us at inf o @lumi l eds. com lumile ds li ght i ng, ll c 370 west trimb l e road san jo se, ca 9 5131 ?

▲Up To Search▲

Price & Availability of AB20-6

	To Download AB20-6 Datasheet File
If you can't view the Datasheet, Please click here to try to view without PDF Reader .