Please let me know if there’s a better PCB design forum/site you’d ask this type of question? This isn’t directly related to a V1 project, but I know the community here will have good suggestions.
Is it a good or bad idea to use multilayer GND pads, to 1) connect to ground plane, and 2) also be a ground pad for SMD LEDs? Wondering whether I’m risking poor connection, and whether solder could wick into the multilayer pad’s hole? See pic, left out copper pour so easier to visualize.
Also, this is my first custom PCB board, so general constructive ridicule is also much appreciated!
Am using EasyEDA to update my dimension constrained 2 layer PCB that’s crammed full of WS2812B-MINI 3535 RGB LEDs. Cramming as many LEDs as I can around the perimeter.
This v2 revision has 0.6mm or larger traces where possible. Reason is my v1 boards (assembled by JLCPCB via their SMD service) used 0.254mm signal traces, but they don’t work… Observing flickering, and non responsive LEDs, fault location(s) varies from board to board. Many LEDs have non functioning R or G or B colors, or don’t light up at all but still transmit data to the next LED, maybe… Don’t have eyes or microscope powerful enough to inspect LED’s internal wiring. However, applying manual pressure at different points onto some misbehaving LEDs can momentarily cause them to function. This makes me suspicious of the LED quality. Currently using WorldSemi WS2812B-MINI 3535. I’d try SK6812 but jlcpcb/lcsc don’t have stock for needed size. Contemplating redesigning board with 5050, but layout will be trickier because I can’t change overall board dimension, has to fit in an existing housing.
Thinking my code is fine (?) because my adafruit neopixel strip (with SK6812 5050 shiji lighting LEDs) works fine. However, the Worldsemi LED spec (google translated) mentions “The port scanning frequency is 2KHz.”, haven’t figure out yet if/how that affects how I should tweak signal timing within the Adafruit neopixel library. Fun!
Your question could maybe be stated “Can I get away with a regular non-filled via under an smd device, vs paying for via in pad processing?”:
It’s generally not a good practice. Besides solder wicking in, flux and other materials will off gas if it is not filled properly. The result could be hard to predict. Maybe there is still room to be had somewhere without going to 5050? Maybe going to 3-layer is warranted here?
LOL @ running 0.254 traces everywhere. You definitely aren’t the only one… I did the exact same thing on the first project I sent in to a board house, and it had traces carrying ~2A max. It worked, but not the best (this was an analog fan controller). Getting all the gerber files sorted can be a pita at first.
I haven’t ever seen the vias in the pads like that. I have seen vias be a tiny bit proud of the surface, but not on professional boards.
The ground fill should let you put the via almost anywhere, right?
I made an esp led board myself this summer. It was a lot of fun. Feel free to check out the easyeda project for it:
Discussion:
I am pretty sure the BOM was accurate, but I did the soldering myself and I might have used a couple of parts from the bin.
I used WLED, and I can’t recommend it enough. It works a little better if you use their pin recommendation, but I think it can do any pin (but some aren’t as reliable).
The data lines should be able to handle tiny traces. I ended up using a ground pour on one side and a Vcc pour on the other, to make sure power got everywhere.
I have a cheap USB microscope I use for inspection. I have seen tech use those headband magnifiers to great effect as well. My cell phone camera is actually pretty decent at seeing better than me too.
But ultimately, I have to trust the beeper mode on my multimeter. It will tell the story of who is connected to who.
I’m fine talking about it here or there, or another topic on wled for your specific build/issues. I don’t check that forum as much, but I did get an email. If it seems like I am ignoring you, just ping me and I will make sure I check it.
Short: Use jlcpcb’s “Standard PCBA” assembly option with 240 reflow temp when using WS2812B LEDs. Explicitly check datasheet reflow temps being used by assembly service.
Long:
Sharing incase someone else here is playing with LEDs/WS2812B’s and using jlcpcb assembly service…
Fixed bunch of code and design issues on my part. But I couldn’t understand/explain this…
Until now…
Just realized jlcpcb PCB assembly service, by default is “Economic PCBA”. That service lists the reflow temperature as 255.
However WS2812B datasheet reflow peak is 240 So, guessing many of the LEDs were overly cooked.
Fortunately jlcpcb assembly have a “Standard PCBA” that you can optionally select with reflow temp of 240. So, will explicitly use that next time…
Lesson… RTFM, read the datasheets, read the service details. Read everything with the same diligence you’d give to ensuring grub screws are tight.
Ideally, each part sub-component would be tested before assembly. No idea if any Fab house does this, wouldn’t expect from jlcpcb given their low assembly price, still…
So, guessing many of the LEDs were overly cooked.
