Casio Watch with Bluetooth (nrf52832) Part 2

9/2/2023

The PCB design I created in the last part had many issues which led me to do much research into high frequency designs. The book Right the First Time by Lee Ritchey was especially helpful in teaching what to consider when working with high speed designs, which I very much was because the antenna section is aiming to create a 2.4GHz signal. To simplify, any given trace on a pcb acts as a capacitor who's other "plate" is the ground plane. This means high frequency signals will make their way directly to the ground plane, hence a "return path" on the ground plane which is exactly the same shape as the trace. There is also a wealth of information regarding impedances, so I handedly recommend reading at least the first part of the book if you are designing any PCBs with microcontrollers running at more than a MHz.

However, the real mistake with my last design is that I somehow missed the reference pcb offered by Nordic themselves. Sadly the format it is in can only easily be opened with Altium Designer, and I am not nearly ready to invest in that. The tool I was using, EasyEDA, could not open this format at all, so I switched to KiCAD because it could. Though it was messy, the folder structure is non-existent, the improved routing tools more than made up for it, and using the reference design sped up the process significantly.

I failed to consider the antenna reference design in the last part as well, but this time around I (mostly) followed it. It made squeezing in all the parts on such a small footprint a challenge, and I had to remove half of a millimeter of the antenna's section horizontally. Hopefully the signal will be sufficient to reach my phone from my watch.

I've always dreamt of a reason to finally buy a rework station, so this project was the ultimate excuse. I settled on the wonderful Atten ST-862D recommended by Louis Rossman with which I have no mechanical complaints after having used it for about two weeks. However, it was pretty problematic to use it because components were getting blown off and the solder seemed too stubborn to melt. As with most things, it looked much easier on youtube. With practice I got better, but I am still worried. I don't know when things are working if the chip is burnt or if it's some other issue. Such is the process of hardware debugging.

I got the board all soldered except the antenna section which I will get to after I can get basic functions working. It was going well until I plugged in the programmer incorrectly. I swapped the 3.3V with clock (I am not using a header because of space constraints), which caused the setup to warm up quite a bit. After that the programmer stopped being recognized as a valid USB device, so now I'm waiting on the replacement(s). I attempted to use an arduino in combination with the Arduino DAP library to perform the programming but it is not detecting the chip and I don't know if it is an issue with the programmer or the board, the whole reason I tested the programmer before. I did replace the chip, which did not fix the issue, and after which I also noticed there the arduino is supplying a 5V signal (D'oh). I'm not sure if that is enough to kill the chip but it would be a shame. When the programmer arrives I will know regardless, and will likely have to solder a fresh board.