Hello everyone,

I'm just looking for a sanity check. I want to drive a 5V LED strip using one of the RGB outputs of a 5V WS2811 module.

I found this discussion, which is for a 24V supply using a PNP transistor:
https://forum.allaboutcircuits.com/...using-ws2811-to-switch-pnp-transistor.203523/

The WS2811 has three output pins. Each output is a PWM controlled constant-current sink (about 18mA). An LED is connected to 5V and the WS2811 output. When the PWM is "on" it allows 18mA to flow through the LED. When "off" it does not allow current to flow. The LED is usually one third of a single RGB LED package, but it's easy enough to use the three outputs to drive three discrete LEDs. I want to use a daisy-chain of WS2811 chips to drive some individual LEDs, and some 5V LED strips of maybe 20 LEDs. It is convenient to use 5V everywhere, but a WS2811 can not drive an LED strip without some sort of buffer.

There are several examples online using MOSFETs to drive long strings of LEDs, with an inverting stage (because of the inverse sense of PWM switching versus LED "on" state). If everything is 5V, and current requirements are limited then a single PNP transistor ought to work.

So here's my design.
The 4k7 resistor pulls the transistor base high when PWM is not driving. The 180R resistor allows the full 18mA of current to flow through the WS2811, and fully saturate the transistor. The transistor is rated at about 600mA, which is sufficient to drive about 30 LEDs in a strip at about 20mA each, although I intend to limit my use to about 20 LEDs.

I have built this, and it works, but I have not measured any voltages or currents in the assembled circuit. I'd like to hear opinions on the design, especially if there are any ideas to improve it. The end goal is to control several individual LEDs and several "illuminated areas" across a panel about A4 size with a single 5V supply and a single control interface.

Thank you.

 

did you read the datasheet?
https://cdn-shop.adafruit.com/datasheets/WS2811.pdf

WS2811 can drive multiple LEDs directly if using higher voltage.
connecting LEDs in parallel can work but it is a bad idea - unless you provide individual current limiting resistor per LED (which you do).

the next issue is drive current. to get 600mA output current, that PNP will need base current 1/10 of that which is 60mA. unfortunately that is more than 3x higher than WS2811 can do. so you may want to buffer it...

 

Thanks for the reply. I have read the WS2811 data sheet, and I understand I can drive a series string of LEDs from a higher supply voltage, but I want to drive a strip of parallel LEDs at 5V. Strips like this are off-the-shelf items.

Regarding the base current, I read that the 2N2907 has a typical gain of 100. The datasheet gives a range of 50 to 100, so I used 50 in my calculations.

 

I would suggest replacing the PNP transistor with a P channel mosfet.
AO3401 comes to mind.
The 180 resistor can be eliminated as well.

 

I would suggest replacing the PNP transistor with a P channel mosfet.
AO3401 comes to mind.
The 180 resistor can be eliminated as well.

Thanks! That's exactly the kind of suggestion I was looking for. I'm not really up to speed on MOSFETs, but I'll sketch up a design tomorrow based on this information and see what you all think. I can source AO3407 locally, but not AO3401.

 

The AO3407 is fine.