The forward voltage is intrinsic. It varies from one led to another but does not vary for the same led.
It is what it is. What is it ? 3.0 V or 3.6V ? Do you know how to tell ?
The led will NOT come on (AT ALL) , UNTIL it reaches the forward voltage.
At that point it begins to conduct and you can see it. That is the voltage you need to measure and record. Nothing that happens AFTER it has begun to conduct has any bearing on the forward voltage. It is either conducting, OR NOT conducting. If you have NO MEANS of varying the voltage and ramping it up gradually to determine the forward voltage empirically , then so be it. If you do , please do it and post the forward voltage. Vf = 3.0 - 3.6V is a statement that applies to millions of leds, not one. Within a large group the voltage is not the same for every led. It should be close though. Please find out at what voltage the led begins to conduct and post that. If you have the datasheet or a link please post that. The Vf is a rating given at a specific current, meaning it draws THAT current at THAT voltage, so Vf= 3.0 to 3.6 V at 350 mA does not make any sense because it draws more current at 3.6 than it does at 3.0V.
BD135-BD137 sounds about right.
Agreed.
Based on the VCE for the BD137 transistor (VCE= 0.5 V @ 500 mA=>0.303 V @ 0.303A,
LED Vf of 3.3 V (0.303 A * 3.3 V = 1 W)
VCE = 0.303 V
VR = 0.7 V (0.303 A * 2.3 ohms)
Voltage drop across transistor and resistor = 1.003 V
V+ = ====> 4.303 V
Conclusion: For the Constant Current circuit given, you need 4.33 V for V+ because the voltage drops across the transistor and resistor drop 1.03 V.
raschemmel: Conclusion: For the Constant Current circuit given, you need 4.33 V for V+ because the voltage drops across the transistor and resistor drop 1.03 V.
Now note - this means at the very least 4.33V.
You do not go looking for a source of 4.33V. To make it work, you use a 5V supply of some sort - which is of course, readily available (but of course, not from the Arduino 5V pin!).
Is there a reason why you wouldn't just use a single chip current regulator, if you're going to use a linear regulator? The AMC7135 will do the current regulation for you (it was designed specifically for driving 1W 350mA LEDs). Dropout voltage is very low, too (lower than the datasheet indicates, iirc), like .2 or so. They're dirt cheap too.
Also - LEDs with 350mA current rating are sold as 1W on ebay, regardless of their forward voltage spec which is often inaccurate anyway (I have seen 2.8v quoted for a LED with a forward voltage of 2.0 at the rated current, and dual die LEDs with the specs as if the dies were in series when they were wired in parallel). I always test mine (with a current limiting desktop supply), in the assumption that all the specs I got were bogus.
No I don't, and I didn't suggest anyone look for one. I simply stated the voltage required. How they achieve that is up to them. I have a variable voltage, adjustable current limit bench Lab Supply. Someone else would need a buck converter or an LM317 regulator.
