Frustration and Delay
It has been a frustrating few weeks with very little progress and time ticking away to the deadline. First, my computer crashed 4 days before I left for Europe. The new computer was mostly set up before my flight, but I got very little else done. I thought this trip would be an easy visit with lots of free time. When I got there they had an itinerary for me every day from 8:30 in the morning until 10 pm and I usually had a few hours of work to do in the hotel after that. It was a great trip from a professional standpoint, but not what I was expecting. The second week was vacation with my wife. I didn't expect a lot of computer time while we spent Thanksgiving week in Munich. There is a reason November is not high season in Southern Germany, but Castle Neuschwanstein is beautiful in the snow and the hordes of tourists were non-existent.
Their was the inevitable 'catch-up' at work when I got home, so I only had a chance to get back into the project over the weekend.
One of the reason I jumped into this project was to learn about and use e-paper displays (EPDs). I find them so much easier to read than LCD or OLED. I guess the luddite in my me still wants to read paper, not a glowing screen. The downside is that they update MUCH slower and the interface is more complicated. The EPD will be an excellent display for my remote sensors (indoors) but may not be the best choice for the main controller. It may not be able to update fast enough when the user is changing the settings. I hope I can find a work-around such as only updating part of the screen or skipping some of the steps in the update process and accepting some ‘ghosting’ until the user is finished, then do a proper screen update.
I chose the Pervasive Displays panels, not because they were the best choice, but because they appear to be the only e-paper choice for hackers. I wrote my own driver to understand the interface, but so far it isn't working (Grrr). My first step was to verify all of the IO twiddling (that’s a technical term).
The interface appears to be correct so there must be something in my SPI setup. With luck I’ll get that working in the next day or two and can move forward. I owe a salute to repaper.org and Adafruit for their interface board, library and tutorial. I didn’t use their files as a starting point (they are for the Arduino and MSP430), but when mine didn't work I compared code to make sure my interface is behaving the same and used an Arduino to verify my EPD is working (it is).
I am not giving up hope yet. I should have a lot of time over the holidays to get the overall software working and complete the circuit board design and layout. It will be really tight to get it working by the deadline, but I’m still in the game.
One thing went well for this project. Every time I use LEDs I have to verify the Cathode and Anode and I still get it wrong sometimes. I had an epiphany while working on this project. I put a red paint pen mark on the Anode side and a black paint pen mark on the Cathode side on the LEDs in my breadboard kit. Whoopee! Now I’ll always know which way they go.