28 May 2025

A correction from last entry, tying the HOLD and WP high will disable them. This means that the chip's pcb is designed to always be in write-permissible mode. Obviously the projector does not overwrite the chip each time it plays, so it must simply send read commands exclusively, and so can I! This makes it much easier to overwrite. I attempted to attach jumper wires to the pcb leads with electrical tape - no success, as the wire leads kept slipping off the pcb and not passing a continuity test -- no matter how much tape I put on. The electrical tape also left a residue on the pcb, luckily nothing too bad. If it had got bad, I would have used isopropyl 99% on a Q-tip to remove it, should be safe for electronics. Next I tried hot glue as a suggestion from ChatGPT. It almost worked! I tried hot gluing the leads on the pcb as if I was soldering them, but the glue would seep under and insulate the leads. So I put a bunch of hot glue directly on the flat surface of the pcb and let it harden. Then I used the leads to pry a hole under the sheet of glue hoping that the glue would keep the leads on the pcb. However, since the pcb leads were so close together the whole flap came loose, except for the top closer to the actual chip, so it would not exert pressure on the wire leads. I tried to electrical tape over the glue flap, but there was not enough pressure. I then used the alligator clips on my mounting stand, which kinda helped except it would not apply equal pressure, so the distal wires would still be loose, even when I used two clips from each side -- the middle would not contact. Defeat for the night. The next morning, I had a thought to use the SOIC8 clip (that I had thought was useless) directly on the chip (not the pcb) -- embarassingly, it was very clear that this was the intention of the clip the whole time. I thought I would be successful until I did a continuity test which revealed that the leads in the clip were not making contact with the chip pins, no matter how much glue debris I cleared or how far I pushed the clip -- the chip was just set too deep in the pcb. A whole day of brainstorming with ChatGPT led me to the idea of using an edge adapter. I think I will eventually get one since I have 2 more chips to dump and modify. I feel this is the most non-invadive way to connect it. I also thought about what the projector uses - ChatGPT did not recognize the pcb as any standard form factor, so the chances that the adapter the projector uses is proprietary. So that lead me to the final resort -- soldering. I did not want to mess with the board in this way, however I do have a spool of solder wick, and the pcb surface is flat, so I can clean it up nicely. This is the first thing that I have ever soldered, but it turned out functional! Initially, I had equipped the tiniest solder tip so I could be precise. Unfortunately, the distal point of the tip never got hot enough to ever melt the solder, even on max heat setting. So I set it down and let it cool so I could switch the tip to a larger flat edge tip. This worked way easier -- the tip edge of this head was melting solder well, and I soldered the rest of the wire jumpers on the pcb. with each solder, I would perform a continuity test with the multimeter to ensure the other end of the wire connected with its corresponding pin(s) of the chip itself. All soldered wires have passed! Now I can finally hook up the pcb to the uno and run the scripts. I also had ChatGPT update the scripts for the uno, and write some overwrite scripts too. I have uploaded the new dump scripts, and will upload the others tomorrow.
The next step is to reduce the 5V of the uno to 3.3V for the p25d80sh. For that, I can use a series of resistors to make a "divider" on a breadboard. I used a top resistor of 2k ohms, and a bottom resistor of 3k ohms (made from chaining a 2k with a 1k). This setup should reduce the voltage to 3V, which is operational. The chip needs at least 2.6V to turn on pins and be functional (up to 3.6V), so 3V is right in a good range. These are seen in the photos with the soldered pcb. I will have to step down the /CS, MOSI, and SLCK pins, then use the 3.3V port on the uno for the triple-tied VCC/WP/HOLD pin. I will also have to learn how to get that script onto the uno so it works. Soon I will have an actual .bin file! So close now.

---

⟵ Older	Table of Contents	Newer ⟶