preliminary WS2812B 8x8 Matrix Booster Pack

[JWoodrell 285]

I layed out the initial planning for an 8x8 matrix using the WS2812B LED's Rob posted about recently.  There is a 2.1x5.5 barrel jack to provide power cause 64 LED's will draw a lot more than a USB cable can provide (almost 4 amps at 60ma each).  so I provided a 3.3v regulator to power the launchpad's VCC line.  there is a single decoupling cap shared between every other LED since the power pins are next to each other(pretty close), should still be enough. half of the LEDs are spun 180 degrees to make routing the data lines nicer.

 

I haven't worked with these before so I don't know if a string 64 long will be an issue for the LP...i don't know the timing.

 

right now it just takes P1.0 to the data in of the first LED, and daisy chains the rest.

 

 

thoughts, ideas?

 

Rob said he can get these LEDs for $0.13 each at >100 so its only 8 bucks worth of LEDs... although soldering will be umm difficult without a one pass reflow setup.

 

the tops of the header legs will need to be clipped so they're flush  but I dislike SMD headers.

 

 

 

led_WS2812b.zip

[multivac 19]

you most likely know this, but if you plan on soldering those with an iron, try to get the pads to stick out of the footprint a bit, so you can heat them with the iron at the same time as the pins. just a little detail i oversaw that made things complicated.

[JWoodrell 285]

the problem I had was trying to keep it under the 5cmx5cm size that is inexpensive from SEEED and or ELECROW.  you should be able to solder with an iron in theory with a sharp tip...  the leads do go up the "side" of the LED a bit. 

[spirilis 1,264]

How do the boosterpack headers go on? Looks like they're under the footprint of some LEDs... SMD pads on the bottom would help though.

edit: nm, didn't see the bottom of your post

 

Sent from my SCH-I605 using Tapatalk 4

 

[RobG 1,891]

Here's how 7x7 would look like

 

[JWoodrell 285]

changed it over to SMD headers and had to change the LED orientation scheme to keep the feet from hitting the ground vias.  the incoming power will be close to 4 amps when all the leds are lit up so it is a nice FAT 80 mil power trace to feed it down to the 5v net on the board.

 

each row going up is flipped around rather than each column like I had last time

 

worked to keep all the ground return paths clear

 

I don't know if I should use a 4 pin header (2 pin 5v+, and 2 pins for GND) although the pin headers are generally rated at 1amp per pin, i can't see using 4in and 4out

 

I also have the option of using a resistor for the VCC voltage rather than the LDO like i saw on @@RobG's board...  so either could be used. (unless I misinterpreted what it was )

[RobG 1,891]

I also have the option of using a resistor for the VCC voltage rather than the LDO like i saw on @@RobG's board...  so either could be used. (unless I misinterpreted what it was )

 

It's a jumper which allows LEDs to be powered from LP's 3.6V. You will not be able to get full brightness when all LEDs are on, but it's good enough for testing.