A comment about these super LEDS: they belt out a horrific amount of RFI, in amounts that can't be contained unless extensive shielding is used. It was one of the most disappointing contract I ever had- it simply = could NOT be made to pass FCC radiation specs. --Bob G=F6khan SEVER wrote: >I have confirmed that the correct Vf bin code ranges in 3.51 - 3.75[V] for >my white Luxeon III Emitter power led. I have to confess that my mind a >little bit confused at this stage. After taking account the proper Vf value >i decided to use max Vf value in my calculations (or do i have to choose t= he >typical 3.6V value?) and based on the >"DS45 Luxeon III Emitter" datasheets' page 12 "Forward Current >Characteristics" figure; "i power the luxeon 3 white emitter with 0.7[A] i >can get max normalized luminous flux". So i chose 0.7[A] as a second >parameter of simple equation. That makes R =3D (24 - 6x3.6[V]) / 0.7[A] >=3D 3.4[Ohm] (i will chose proper approximate R value for that) The >dissipated >power will be ~2.5[W] in this configuration. I think i have to lower the R >value a-bit for maximum capacity. I will try my simple experiment tomorrow >morning. Do you think the result predictable, in other words do we say that >we get stable 0.7[A] current or at the start we get 0.5[A] and increasing = to >0.7[A] slowly? > >Another question for red color Lux3 power leds; Whats the wiser solution to >get the best current / normalized luminous flux trade-off? ( If i want to >get maximum flux i have to drive the led at 1400[mA] but that will cause >overload on dissipated power and if i drive the circuitry with avarage >current i will only get half of available radiation. Whats wrong with tha= t? > >I've lost a bit in this lighter area. Hope you clarify the dark areas... > >2006/2/23, William Chops Westfield : > = > >>I think you have fallen victim to two pieces of misleading data. >> >>First of all, the Vf spec of an LED is somewhere on the "knee" >>of the voltage applied vs current conducted curve. If you apply >>a current source whose max voltage is exactly Vf, you'll be on >>that knee instead of in the "current limitting" range where you >>want to be. Your supply of 24V only allows 0.1V "extra" for >>each of your LEDs, which may not be enough to get them "ON" all >>the way. (I'm not saying this QUITE right; we might get into >>another "voltage across vs current through" debate.) >> >>More seriously, you're interpreting a Vf spec for TYPICAL >>performance as what you'll actually see in practice. If you >>look at the Luxeon document "Luxeon Product Binning and Labeling" >>(), and in particular the >>section of Vf binning, you'll see that a white Lux III CAN have >>a Vf as high as 4.47V (or as low as 3.03V, which is quite a range. >>(the datasheet has the full range even for 700mA)) Even if >>you have the "middle" bin ("L") you could have Vf as high as 3.99V. >>If you have a full reel, you can check which bin you have. I don't >>think individual emitters are marked, though. If you have even a >>couple of emitters in a higher Vf bin, you're going to run over >>the 24V of total Vf you can get away with (or at least be even >>further in that "knee" that I was talking about.) Also, you >>junction temperature isn't going to be 25C at that sort of drive >>(although this appears to be helping you; thus the warmup to >>higher current.) >> >>Easiest solution: can you tweak the power supply output up via >>some adjustment or remote sense capability? (OTOH, I think >>1A current is pushing a Lux III quite a bit. That's close to >>4W even at typical Vf...) Note that each instance of your >>light bar will need separate tweaking. >> >>The rather broad range of luxeon characteristics seems to be a >>bit of a sore point among the lighting crowd. Apparently you >>have to be really BIG customer before you can ask for particular >>bins instead of taking your chances... >> >>BillW >> >> = >> -- = Note: To protect our network, attachments must be sent to attach@engineer.cotse.net . 1-520-850-1673 USA/Canada http://beam.to/azengineer -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist