R&D on LED Drivers controlled via PWM.

Right so as some of you may know I have been developing some high end PWM controlled LED drivers for use in our new Evione Lightsabre ver 1.0.

I have purchased readily made LDD drivers as well as played with various Mosfet and Transistor type drive circuits but today I am going basics.

I am going to take at atmega328 AVR, program it with the Control program for you my light (and yes it is arduino based).

Then I am going to use the pwm output to drive a mosfet which in turn will switch and vary the power to the led’s. Hold thumbs.

So here goes the pics. I really need to set up a video cam….

In the above image what I have done is pretty much connect up the bare ATMEGA328-PU to my board. Ignore the Arduino and scope. They are just there for pre beta work but I always take my Arduino projects off Arduino boards and build my own. Saves one loads of money.

Below is the schematic diagram of what I have done here.

If you have a look I basically place the N Type Mosfet into the power circuit of the LED and control it using the PWM output from the MicroController. This is a very crude way of doing things but this is for R&D Purposes. The final product is always more finished.

 

I am running the main DC power out at 20 volts as I am using 2 x 12 volt 900mA 10Watt LED’s as can be seen above. I am running them in series so that the current running is the same through both devices.

Upon powering up the ATMEGA you will see that the LED’s will turn on as the PWM is opening up the mosfet and allowing the leds to get enough power to turn on.

With the LED’s just turned on you will see we are running at 20Volts and 20mA. At present the PWM input into the mosfet is at its lowest.

Now I have programmed the ATMEGA to allow me to vary the PWM output in 25 steps by holding or pressing either an up or down button. In the Protoboard I just short out where I need to.

I up it to about 30 percent of its max. Please note that I will not push it to the max as the LED’s are not heatsinked and therefore could suffer some serious heat damage at full fwd current.

Its hard to see how bright the LED’s really are as the camera adjusts for the extreme brightness. I will setup a camera on manual exposure to capture this as a later stage.

As you can see the current is now sitting at 460mA @20V and the PWM signal is at about 30 percent. The LED’s really are stupid bright already.

Although we did not keep the LED’s connected I did push the MCU to full PWM output which can be seen on the protoscope above.

My next step in this project is to work out the current limiting and build the current limiting circuitry into this to limit the max current to 900mA. Should be interesting.

Until next time happy inventing.