# Building a LED driver

Discussion in 'LEDs and Optoelectronics' started by john2k, Jul 6, 2013.

1. ### john2k

185
3
Jun 13, 2012
Hi Guys,

Up until now i've been using a 12v buck type led driver to drive and led unit with 3 led's wired in parallel. It gives my LED unit 3.2v fixed voltage and a constant current of 300mA. I cannot find any buck type drivers that give lower than 300mA and is a 12V DC input. So i'm wondering if anyone knows how I can build a simple circuit just to give me a fixed output of 3.2v and a fixed current of 75mA. I dont need it to be complex or adjust depending on the if more led's are added or if it's change from parallel to series as this will not be the case. So just want to deliver the exact specification this led unit needs

Thanks

2. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,496
2,837
Jan 21, 2010
Generally there is a resistor on these boards that drops 0.1V or similar at the desired current.

I modified some to give around 100mA for a project recently.

It's also generally a bad thing to use LEDs in parallel.

Show us a picture of your driver and I'll see if I can point out the current sense resistor to you.

3. ### BobK

7,682
1,688
Jan 5, 2010
You cannot have a fixed voltage and a constant current at the same time. LEDs are current driven devices. They are best driven by a constant current source, which will adjust the voltage automatically to get the desired current. One of the simplest ways to do this is with an LM317 voltage regulator, used in a constant current configuration, which can be found in the datasheet.

Bob

4. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,496
2,837
Jan 21, 2010
As an example, this is very similar to what I bought. (I think mine may have been 700mA, but that's not particularly relevant)

As Bobk has noted, for a current source the voltage can vary -- however it will possibly be close to 3.2V for a single LED (or *gasp* LEDs in parallel).

One of the resistors on this board (you can't see it in their photographs) sets the current. This is simply removed and replaced with a larger value which will drop 0.1V -- from memory -- at the desired current.

In my case I wanted about 120mA and as I recall I ended up using 1 ohm -- which must have given me 100mA -- close enough.

The original resistor is a surface mount device, although in testing I used a carefully crafted leaded component. For the units I made up for real I used a replacement surface mount resistor -- so either can be used.