I am new to the arduino universe and have a few questions. I want to power a 11V led with a 7.4V battery pack and a arduino to control it. I think I need a boost converter in order to reach 10V? But then a also would need a current limiter, which would just lower the voltage? At least that was my impression when I watched Great Scotts videoabout it. Should I do it like that or do I have a completely wrong approach?
what kind of a 10V Led is that? please post the link to it, or even better the link to the data sheet of the LED, so that we can see the technical data of the LED. This is the only way we can know what exactly you need to make the LED shine.
Oh yeah of course, i forget about that... Here is the datasheet
And it's actually 11V, I will edit it.
The forward voltage of a LED is a (more or less, depends on current) fixed value so the resistor does not lower the voltage. You need to limit the current through the LED and that is what the resistor is for.
You need a higher voltage than the 11V to be able to limit the current; e.g. 15V. See below image
You now need to calculate the values for the resistor.
If you want the 1A current (max brightness), the Vf is actually 12 V (and not 11 V, see first graph on page 18 of the datasheet ) and there will be 15V - 12V over the resistor. So according to Ohm's law (
V = I * R), R will be 3 Ohm.
The power dissipated in the resistor is the voltage over the resistor multiplied by the current through the resistor so is 3 W; you will have to pick a resistor that at least has that power rating (the ones that come with e.g. kits are either 1/8 or 1/4 Watt and not suitable).
If you want to limit the current to approx. 200 mA, Vf is about 11 V. So the calculation will result in 4 / 0.2 = 20 Ohm and the ower dissibation will be 4 * 0.2 = 0.8 W (rounded 1W).
Take the worse case scenario (3 W) !!
If you power with 13 V instead of 15 V, you're better of as the numbers for e.g. 1 A result in 1 V over the resistor (so 1 Ohm) and 1 Watt.
Lastly, you can not drive this directly from an Arduino; you will need a logic level N-channel mosfet between the LED and the GND.
The thing about high power LEDs is they get hot. This caused two problems.
Heat can damage them, or significantly shorten their lives. You need to provide a heatsink to draw that heat away from the led. A large copper area connected to the anode or cathode, and ideally this should in turn be thermally connected to something large and made of metal.
Temperature changes the forward voltage of the led. This means that the resistor value you choose based on the forward voltage at room temperature may not be appropriate when the led warns up, allowing too much current to flow and you get thermal runaway. So you may need to think about using a constant current driver which will replace the series resistor and will automatically adjust the voltage it drops to keep the current through the led at the desired level.
Thank you very much.
Ok, but if I would use my Arduino (and a current sensor) to measure the current flow and if the current gets too high, could I just lower the voltage in order lower the current or wouldn't that work?
Yes. But since the non-linear nature of the LED wants to hold the voltage across the LED approximately constant, most of the "change" will be across the resistor.
Normally, high power LEDs are driven from a constant-current source rather than using a resistor for current control/limiting like you do with "regular little" LEDs. But it's not an easy thing to build and you probably won't find one that also boosts the voltage.
...Between the heatsinking & current control, high-power LEDs are NOT that easy to work with. The heat issue is somewhat ironic because LEDs run cooler than regular light bulbs, but regular light bulbs are made of tungsten and glass which can stand MUCH higher temperatures than a silicon chip.
I think that is more expensive and complicated than a constant current driver.
Let alone the LED chips (which are not silicon!).
The concept of incorporating an Arduino (or any general microcontroller) into a feedback loop regulating a power supply is singularly unwise!
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.