I am in the process of wiring up the RGB pads located on the bottom of the surface mount version of the particle photon. Here is an image for reference:
I am under the impression that this is used to drive a RGB led elsewhere (on a box enclosure for example).
From looking at the recommended wiring schematic, I am assuming that the Photon is switching a ground to control the LEDs (PWM), can anyone verify this?
In this case, I am assuming that it will only be able to drive a single RGB LED before the Photon circuit is overloaded is this correct? And if that is the case, is it feasible for me to create a “high side switch” out of a transistor and mosfet, have the photons switching ground control that switch, and run my own LED driver that can handle higher current levels? Thanks and any help is greatly appreciated!
Unless you use a current hog of an LED, I think you will be OK just wiring in parallel - with the 1K resistors there, we know that each leg of the on-board LED is only drawing something like 1-2mA, so there should be plenty of drive left on those pins from the STM32.
If you wanted to buffer it, with outboard circuitry, that is certainly quite doable. Off the top of my head I do not know if the pins used to drive the LED are open drain or push-pull. If they are push pull, you could probably use a CMOS buffer/driver rather than a discrete solution. If open drain, you may need to add a high value pull up to your external circuit.
Hi Andy, thanks for the reply!
The LED being used on the device is going to be anywhere from a hefty single RGB LED, to a RGB LED strip (less than 250mA). Because of this, I figured I would implement a LED driver of some sort so that it could be flexible on how much power it draws. I have never used a CMOS buffer/driver before, but I’m assuming it would take the place of the transistor/mosfet combo, I’ll look into it. You don’t by chance have a part number or a common part that I could google and look up for the CMOS buffer/driver?
Sure - if you’re running up to 250mA, then you’ll want something more than just a cmos buffer. You were on the right track all along. Consider using an inverter (e.g. 74HC04) to convert the -ve going pulses into +ve going pulses that can drive an N-channel fet between the LED and ground; but there’s a zillion ways to skin this cat.