SN35 Mini Project Update #03


It's been a week since my last update.. my bad. It's been quite a hectic week though trying to establish whether the project is actually going to end up being feasible or not. The current verdict? Cautious optimism. Ask me again tomorrow though. Some of the biggest challenges I had this week were:


  • Getting a screen to work at all, and then risking breaking it after I did (spoiler: I'm fairly sure I did, but I haven't confirmed yet)
  • Sourcing parts before Thursday as there are no deliveries on Easter Friday
  • Not being able to get a professional PCB fabricated and delivered because of the above


Keep reading to see how I am attempting to work through the issues. I also had a great chat with both the Ben Heck team and dougw late last night, so I am feeling very inspired to keep hacking at it. With a bit of luck and persistence, I'm still confident I will get a working build and an entertaining video done!


I'm also even more excited to start talking with people from element14 about what a potential YouTube show could look like than I already was before, so start sharing with me what you want to see in future content from potential element14 content partners.



The "Working" Screen


If it's not obvious by now, I've never made anything with an embedded screen before. For me, this will certainly be the "big challenge" of this build (among a collection of others firsts). After my inability to get an Arduino compatible screen working (pretty sure it was a lost cause anyway) – I was hoping for a painless experience with the Adafruit module. However, a few things didn't go right. One was mostly my fault, the others not so much.


But at least the LCD is working for now. Yay.


If you ever build a project using a PiTFT screen and wonder why you can't get it working, it could be that you are following several seriously outdated tutorials. For reference and as of early 2018, this tutorial from 2015 seems to mostly work! The driver is already in the kernel, and you don't need to manually configure the device tree which is nice. The other thing that took longer than I care to admit to solve, and was mostly my fault – check the connections are aligned right. But also check you are using the right header. Duh. In my defence though, if you are going to have two available 2x20 pin headers, why would you completely change the orientation of one of them...?


A big gash. Still waiting to find out the result of this, but it probably means the touch screen won't work, of that the entire thing becomes very flaky, or that I finally caught a break in regards to my break. For now, it still turns on and shows an image at least.


There is only one way the LCD will fit in my proposed design, and it's sure as anything not going to be on the PCB it arrived on. This meant it had to come off (no footage because I was fairly confident there was going to be profanities if it went badly). I very carefully attempted to pry it off and was worried the glass would crack. It was going very well until the last bit of tape gave way, and the thing flew off – right out of the ribbon connector. The glass is fine, but part of the ribbon cable tore. I'm reasonably sure that the damaged area is related to the capacitive touch screen given that it still turns on, and from what I know after having looked quite intensely at the data sheet before hand.


Kind of ironic though since I spent a bit extra to get the capacitive over the resistive touch, right? For now I'll just keeping going and will have to skip the touchscreen if it doesn't end up working.




Professional PCB Alternative


After being up most of Sunday night working on a PCB so I could send the design of to be made on Monday, I realised there was no chance it would arrive in time due to the public holiday. So after being very sleep deprived, I went to bed way too late which has really put me out of sorts these past couple of days (hence the lack of updates). Lesson learned, don't stay up past 1am no matter what needs to be done. Something something.. cognitive function good. Something something... sleep deprivation bad.


No sleep isn't working well. I wonder if the beer thing has something.. it apparently stops Homer from going crazy. Maybe the Ballmer Peak works for hardware prototyping as well!?


At this point, I think the safest thing to do is to etch a few smaller boards and wire them together Ben Heck style. Smaller DIY boards tend to etch better than larger ones, which is especially important when trying to get down to a pitch of 0.05mm. On the positive side, I've etched boards before. I may use this part of the build as the "teachable moment" and quickly walk through the process of etching a PCB.




Not To Jinx It..


Two schools of thought on how to acquire a parts library. One is to buy a few extra parts at a time based on each project you build. The other is to bulk buy everything up front. Some how, I've ended up doing both in one shot?


I've received two boxes from Element 14 this week. One very large, but "free" box thanks to some awesome contest I entered. And then there was one smaller, but more expensive box. As it turns out the larger box was missing a few crucial parts that I needed for my build, but I've found a local supplier who can express post them. It's [was] mainly the Powerboost circuit I am [was] waiting on, so hopefully it will get here in time and thankfully It came while writing this! Hopefully this doesn't jinx it, but in theory that was the last part I needed that I can't buy at a local brick and mortar store – excepting a few IC's that had a 6-7 business day lead time. I might have to pinch them from the existing modules.


Because of the time crunch, the smaller box ended up containing quite a wide selection of parts as I'm still not 100% sure on which components I will end up using in the final design. I really didn't want to find myself short. No doubt I will use the parts for future projects though.


About a quarter of the expense of the small box was for a 10–pack of these filters for my desoldering gun. I'm thinking I may need to use it for the project, and the original filter is very well blocked. I will need to call and ask why they only sent one though!



Hopefully this will be the only "frustrated" post from me for the duration of this competition. However, the 3D printer is not showing me much love either at the moment... The next major challenges are going to be fabricating and soldering a PCB with the fine pitch, and securing access to a suitable printer over the long weekend to make the case. The next post will likely see me wiring everything together breadboard style, and running Retropie on the Pi Zero.


I think these types of setbacks are mostly to be expected when you are tackling so many new things at once though though. Whilst in my mind it seems like there is a lot of bad luck cascading together, it's all part of the journey and I'm excited to keep pushing onwards. It is important that I do push myself quite hard for this competition at least, especially so that I can clearly show that I have what it takes to Hack Like Heck. At the same time, it'll be equally important that I am aware of how far I can realistically and consistently push myself on a week to week, or even a fortnight to fortnight basis though.


Let me know in the comments.. does it help you in any way to hear about the difficulties others encounter? How do you manage to stay positive when things don't go right for a build? Stay tuned, and happy hacking everyone!