dougd wrote:The problem is bad pixels!
At this point I have to agree!
I'm wondering if the failing nodes are mostly TM1804's? The WS2811 being a constant current device only uses a resistor on the red channel and that just to balance the load on the chip. The TM1804 being a constant voltage device requires three resistors and they set the current.
I've seen the mold with the PCB in place and there is not much room on the PCB to get that many parts on the board. If a direct substation was made the TM1804 is going to get overly stressed. We won't know till Steven can open one up so I'm waiting to hear what he finds.
Well as mentioned earlier. I am seeing the exact same things out of my pixels. 97 strings of 50. They are 2811 bullet nodes. At least 50% of the strings have failed. I have a shoebox full of cutout pixels where I have tried to fix the strings to no avail.
The reason I asked is that there are SO MANY variables that can cause what's in the video...sometimes a diagram of how you hooked things up can help. Might be nothing...could be a simple ground issue. My first attempt to fix that would be to disconnect each string and go one by one plugging them back in...I will say this...I had a strip go white, it was a data issue.
dougd wrote:Well as mentioned earlier. I am seeing the exact same things out of my pixels. 97 strings of 50. They are 2811 bullet nodes. At least 50% of the strings have failed. I have a shoebox full of cutout pixels where I have tried to fix the strings to no avail.
Silicone injection molded or epoxy encapsulated?
