Reverse engineering high power LED driver board

R

RichieNorthcott

Member
So I have been trying to reverse engineer these boards I have. These use a 4 channel IC so respond the 4 bytes of data rather than the usual three.
circuit.jpg

Red - 4 x PT4115 drivers and associated components. My labelling is actually a little misleading since the LED negative connections aren?t common - they are also associated with each of the four sub circuits The is a sample schematic in the data sheet for the PT4115 here:
https://www.electroschematics.com/wp-content/uploads/2014/07/PT4115E-datasheet.pdf

I also note these have been used elsewhere in this forum on the ?20W LED driver? boards, which is more or less this.

Green - NPN transistors interfacing the WS2811 to the drivers

Blue - 10K base current limiting resistors for the transistors

Orange - 51 Ohm resistors on data in / data out lines

Purple - 1K resistor to drop 12V supply voltage to around 5V for IC, and mystery component! See question below....

I have three questions:

What is the component above the 1K resistor? It?s an orange cylinder with a green band on it. It looks like a glass-like material

I was thinking about using these on my garden LED spots - but this would mean long connections between the board and the LED which is probably a very bad idea since that changes the overall inductance i.e the driver ICs assume no notable effect from the connecting cables.

There are two capacitors below the IC connected to ground - I assume this is a way of getting more decoupling capacitance i.e. two in parallel?

Thanks!
 
Last edited:
That looks like a diode. It might be trying to protect the data input from spikes. The data lines can be long the wires from the output driver to the LED should be short. Put the board as close the to the LEDs as you can.
 
It might also be a zener diode, which would be used in conjunction with a 'dropping' resistor to provide a more-or-less regulated voltage for the WS2811 components. You might actually be able to read the component name from the part using a magnifying glass (unless the name is on the underside of the part).
 
More...

The ws2811 only has 3 outputs, and the board has four, so I don't think that the 8-pin IC is a ws2811.

The maximum value DIM input to the pt4115 is 6V. This increases my confidence that the part in question is a zener diode, used along with the 1K resistor to drop the 12V input down to some value around 5V which goes to the transistors as well as (possibly) to the 8-pin IC. There are quite a few possible part numbers for that part, some JEDEC standard and quite a few not.
 
Thanks - yes I was looking at some videos about SMD diode types, and agree from its form that is likely a Zener

Re the IC it's a UCS2904B - which works with standard pixel timing protocol timing (ish). I say ish because i have issues mixing thee 4 channel chips and 3 channel chips on the same circuit.
 
You must log in or register to reply here.
Back
Top