Constant current driver

[zzonbi]

The leds work half the time, but at higher current peaks, to get the same flux. What you save in driver losses could well go in leds losses.

It's nice and simple I agree, but you are putting the driver in the backseat...

 

[SirJMD]

r0g3r --> Look interesting - what housing does that LT1676 have?. Ive been working a bit with the LM3407, and its eMSOP - extremly small :/

space --> Ill simulate it when i have calculated the values of the two components. Just need to figure out a model for it. Im quite sure it will work - so does my teacher at the university (electrical engineering).
The LEDs are supplied with 15VAC @ 50 Hz. You cannot drop the inductor, and the reason why is simple. If you turn on the circuit at the top of a sine wave, you will have a massive current surging thru the LEDs, since the capacitor isnt charged!


zzonbi --> Since its AC, yes. When youre working with AC, there's a peak current and a RMS current. Both can be calculated. I will of course make sure the peak current doesnt damage the LEDs. But as far as i see it, it should be a problem here at all.

Hopefully i get some time today, to do some calculations.

 

[space]

SirJMD:
As I said you really should start to simulate this circut.
The output impedance of your transformer will very likely be so high that the inductor is not needed. (Your 15VAC transformer is also an inductor. A rather large one.)

space

 

[SirJMD]

SirJMD:
As I said you really should start to simulate this circut.
The output impedance of your transformer will very likely be so high that the inductor is not needed. (Your 15VAC transformer is also an inductor. A rather large one.)

space

Indeed, it is an inductor - problem is tho, that its a salvaged transformer, so i have no idea about its specs (input and output). Maby i should give it a run in a LCR meter.. not a bad idea !