PWM controller for RGB leds

Hi all
I am trying to design a pwm controller to run some RGB leds for a project I've been working on. Unfortunately, as they are super-bright, I can't seem to find a solution that can give me the kind of voltage and current output that I need. My LEDs have a Vf of 2.5 (Red) and 3.4 (Green and Blue) volts respectively and a forward current of 400 (Red) and 350 (Green and Blue) mA respectively. I am trying to run four of these leds (either in series or parallel, it doesn't really matter to me) and control them with three potentiometers (1 for each channel: R, G, B). I have read over Bill Marsden's guide to LEDs and I think I have an idea of what I need to do but I'm having trouble working out a solution that can handle both the voltage and current that I need. Any advice would be much appreciated.

 

If you have enough voltage than run the LEDs of the same color in series. Then add a resistor with each series string to give the maximum current you want. Drive each series string with a PWM controller such as a 555 driving a power MOSFET.

Since these diodes are fairly high current, if you want higher efficiency then you can use a constant-current type switching buck regulator for each string such as one of these.

 

Run 4 of each color in series..
get 1 of these
http://www.mpja.com/15-Volt-Regulated-Plug-Power-Supply-16A-GME/productinfo/18724 PS/
and 3 x meanwell LDD-350H-W
http://www.jameco.com/1/1/47589-ldd...rrent-led-driver-wire-style-encapsulated.html
they dim via a pwm input so the 555 pwm or whatever.

 

Thank you crutschow, I'll give that a try as soon as I have a chance to go pick up the components. mcgyvr, Thanks for the advice, but my circuit is going to be in a confined space (1.4" I/D abs pipe) and needs to be portable (running off a battery).

 

um... how do you intent to handle heat management?
What battery are you using?
and how long do you think its going to last?

 

Second (or third) the idea of running each color LEDs in series. But if battery life is an issue, I suggest making your own buck regulator. It's very easy to do, just an inductor and diode added to a basic switching transistor, and you no longer need a series resistor. Depending on conditions, I might add a current sense resistor (much smaller than the series resistor would be) and shutdown transistor to each color, just in case of accidents. But making the circuit more efficient helps you two ways--longer battery life, and less heat to get rid of.

 

For heat, if I have to, I'll install a small fan to circulate the air (the ends of the pipe will be open). I'll be running it off of a lithium battery pack, and I know I can't expect too much battery life. If I can find a 12 Wh battery then I should get near enough an hour out of it between charges which should be good enough for my purposes.

 

Do you need it "super bright"? You can always drop the current way down (100ma or less) and still have one heck of a bright "rave stick"..

 

I'm not sure. I'll see how bright they are when I get them hooked up to a power supply in the lab. If they're more than what I need I'll probably just run less of them (maybe two or three).

 

just a tip.. make sure you don't stare directly at the LED's.. They can destroy your vision VERY quickly. they are BRIGHT..SO BRIGHT.

 

Finally got a chance to test this in the lab today and it works out nicely using a current limiting power supply, but the led won't draw any more than 50 mA. I can't figure out why.

 

Gonna have to actually post a picture/schematic of how its wired and what you are specifically using to power it. Does your power supply (hopefully in constant current mode) go higher? or what specifically happens when you adjust it higher

 

I'll get a schematic up soon. As for the power supply, it's a fully adjustable power supply set in current limiting mode (set to 350 mA). No matter how much I increase the voltage, the current won't actually go above 50 mA.

 

Here's a schematic. Sorry it's not very legible; the program I have to work with is not very cooperative.

 

I'm wondering if my mosfet could somehow be limiting the current to the LED. I've looked through the datasheet and from what I can tell, I'm not surpassing any limits, but I don't know what else could be causing this issue. The mosfet I'm using is an IRF2804. Any ideas?