September 30, 2025
Last time, I got the boards and one actually showed up on USB. Since then, I soldered the connector onto the board (the trick is applying liquid flux and using the Pinecil's chisel tip with solder on the rear side), programmed the 60Hz VA and VB/VCOM signals, found a bug in the waveform code (the second DMA was transferring twice as much data as it's supposed to, so it was staying in progress between re-configurations of the PIO), built a stupid plastic carrier, and tested the screen. It doesn't work.
Initially, I guessed that the video signal was malformed, and while this ended up being correct because of the issue mentioned above, it's actually not the problem. Although the DMA problem caused the second frame to send only half a line of data and the third frame to stop running with BCK high, this doesn't matter if you don't wire the board correctly. I apparently never thought to check which pin Sharp labels as pin 1 on the display connector, so I accidentally inverted the connector footprint. I probably fried the display I was testing with, unless it can handle 5V on a random data pin, but the board is just fine. Unfortunately, the only way to fix this is to get a new board made.
The picture below is from the mechanical drawing at the very last page of the datasheet, showing the connector up close. Also, interestingly, the image of the display on the DigiKey listing shows it with the bottom side up, which means that the display cable does not flip around like I thought it did. I don't have a picture in this post, but the back of the screen has a reflector (or maybe a polarizer?) attached to it with a shiny back. This shiny back is what the DigiKey listing shows.
They're not much to look at, but here are some pictures of the rev 1 board and the display. I ordered 2 displays, and DigiKey shipped them wrapped in bubble wrap inside an anti-static bag. You can see the marks left by the bubbles on the surface of the display. I think these will clean off with some isopropyl (what Sharp recommends for cleaning) but I haven't tried yet.
Displays, presumably tenderly wrapped by hand, because who's buying two of these at a time.
I'm particularly proud of this one, which I soldered after discovering the flux trick. I practiced on the non-populated boards too many times before doing this.
Look carefully at the screen. Those round marks on it are where the bubbles of the bubble wrap were pressing against the display surface.
Compare the little dark thing inside the red rectangle to the diagram of the connector above. That's the "plating lead", and it's on the pin 21 side. (This is the first connector I soldered, somehow perfectly by hand on the first try, without flux, which is why it looks blobby.)
Revision 2 cleans up the design. It wires the display connector properly, has a pull-up resistor on the RP2350's reset line, doubles the LED resistor to make it dimmer, takes out two extra GPIO pins to make routing a little easier, removes the 3.3V regulator since every component and power supply can run at 3.2V, adds a resistor on the charge pump output to cause 5V to drop quickly when the charge pump is disabled (currently it takes several seconds to fall, and Sharp says to "please not use extremely slow power source"), and adds board art of my choosing at the pedantic request of a friend. (Future boards, in this project or another, may feature chibi anime women if sufficient resources are available.)
I have ordered the boards and more display connectors---I may have destroyed all but 1 on practice boards before I realized the flux secret---so I can get right back on my feet.