Failure of Rev 1 And Lessons For Rev 2

A project log for Custom Smartwatch

It displays time

Matthew James BellafaireMatthew James Bellafaire 12/08/2019 at 13:100 Comments

It's been awhile since I've updated the logs for this project (and that mostly has to do with the busy semester). In my free time I've been working with the watch trying to get something to work. I built up the board and tried to get it to work to no avail. Turns out in the original design the UART bridge was giving me issues. I've since switched from the CP2103 to the FT230X (which comes in a package that isn't QFN). 

I am still unsure about what the issue was with the REV 1 board. all the power circuitry seemed to work fine (over voltage / under voltage / 3.3v LDO). However no matter what I tried the UART bridge would not be detected by my computer. My best guess is that it has to do with the distance the usb trace had to run (the L shape board made layout fairly tricky). Either way I've corrected as many issues as possible and was much more careful about handling the RF of the antenna. Another issue on the rev 1 board was that I forgot to define the soldermask pullback. This left most of my QFN packages without any soldermask between the pads (I was able to hand solder but it still took far too long). Either way there's a lot of improvements here (that hopefully amount to the board functioning). Here's the new board: 

The header probably raises the most questions. In the first revision I decided to go with a 3 button interface for the watch. I've since decided that I would instead go for the ultimate goal of the project which was a full 1.8" touch screen. I finally managed to track down a touch screen that was the correct size,  (From bangood) and have ordered 2 of them. Once I have them in hand I'll decide whether to order/redesign the boards. For now the project is just kind of on pause since finals are next week (and I'm waiting on parts). If the 1.8" screen is too small I may end up going with (which in general is much nicer but is many times more fragile).

I've decided to go with a single sided load due to how much of a pain assembling the double sided load was previously ( it made applying solder paste nearly impossible for smaller pin pitches on whichever side was assembled second). Either way that's most of the changes, I plan to release all my schematics/board files/BOM when I finally have working hardware to program.

I have a long holiday break coming up and I plan to work on 3 of my current projects then bundle the board orders together. So it may be a bit before I have these in hand. Either way, see you in the next log!