Chris.Elston

Success! Apart from connecting the probes the wrong way round that is... Thanks for your help audioguru. ;D

Will the questions never cease?! :D

I've been playing around with the power figures you gave here for about 20 minutes, but I can't see how you derived them :(

StathisCy, the circuit you posted was my starting point, we've been discussing a solution which is an improvement over that.

Audioguru, would you mind explaining the choice of the TIP31? It seems to be a little overkill for this application (Ic 3A, 2W). Would something like a BC639 do the trick just as well? (Ic 1A, 2W). It also comes down to TO-220 vs TO-92, it's all board real estate :) Thanks, Chris.

Thanks, I understand that now. So if I read right, the 2.2R is a function of Vbe and the desired current, and in no way related to the forward voltage drop of the load or the supply voltage? Presumably if you increase the supply voltage the power dissipation and therefore the component ratings need to go up as well? Cheers, Chris.

Maybe spoke too soon :) Can you explain what effect turns on the bottom transistor? How did you arrive at the 2.2R value (I understand the 0.65V/R, but how is this derived) Is it that the resistor limits any further current, so the current flows into the base? I understand the action of the two transistors together to turn each other on/off to maintain the required current, but I don't understand how that reference current is set.

Thanks very much. I think I'm good to go now :D

:( Because it 'robs' current from the base of the top transistor? Cool, I think I get this now. As I mentioned before, I'm planning to dim the LED via PWM.

Thanks for the answers. Any chance you could take me through the theory of the circuit given? I presume that I'll need transistors with an Ic at least the current I plan on switching. If you can explain a bit about how it'll work, then I should be able to select the correct components. Thanks, Chris.

I guess that people here are probably tired of LED questions, but I'll give these a shot anyway. 1) LED forward drops are specified over a smallish range, e.g.: 2.0 - 2.2V. Does this indicate a dynamic range, or is it that any one specimen of that type of LED will have a fixed drop somewhere in that range. 2) I want to drive a high brightness red LED @ 350mA from a 5V source, assuming a 2.2V forward drop across the LED. I calculate the current limiting resistor as 8R, power dissipation across it as ~1W. So far, so good (I hope :) ). The nearest preferred value resistor available is 8R2, but not many vendors carry 1W 8R2. So basically, the question is, what are my options beyond ordering the aforementioned resistor? I know that there are some ICs out there that do constant current, but I was wondering if there was anything I could do with good old 0.25W resistors to spread the power dissipation across multiple ones. 3) How critical is the current anyway? I know that 350mA is the recommended current, and that they will do 400mA pulsed. I plan on controlling the brightness via PWM switching, so how does, for example a 50% duty cycle affect the current through the LED. My guess would be that you get 50% at full current, then 50% at none - and that the off periods allow the LED to cool. This effectively cooler running means that you can drive it at 400mA in the on phase? Cheers, Chris.