I have two N64 Adv. V2 PCB's I picked up from borti off twitter a year ago 202200729(d) and 3 flex's v20230813 all unpopulated. I recently populated both. While waiting for the n64 i was going to install into I verified per the github that 1.8v 2.5v 3.3v 5v had no continuity with ground. Then I wired 3.3v 5v GRD from another console directly to the manual install pads, both LED's D1/D2 lit solid red/green. I hooked it up to a usb blaster and quartus loaded the newest n64adv2_2p23.pof, I checked all three boxes program and verify (CFMO/UFM) it loaded '100% Successfull' and readout the exact message as the github page photo with the blue checkmark in the 10M16SAE144 FPGA icon. The N64 arrived and I installed the flex, I connected the FFC to the V2 board and checked continuity from the top of every RCP-NUS pin (above the flex solder) to the manual install pads on the V2. Perfect continuity and no adjacent bridging continuity. I did this to verify through the flex solder joint and through the FFC connector on V2 all the way to the manual pads. I feel confident in the flex install. I put a jumper pack and game cartridge in and got not only no picture but absolutely no signal on a gaming monitor, a 1080p tv, or any detectable signal to a retrotink 4k. And I'll mention when i jacked direct power from the first system both led's were on solid, through the flex with the failed signal I was only getting the red solid D1 led, D2 green flashed on at power up and instantly off.
Now I went through the debug algorithm off the firmware github checking continuity from all the points borti suggest: U1, U7, U6, U5, all the D0-D6 from the FPGA to the FCC connector, and the rest of the misc. signals from FPGA to FCC connector. Every signal checked out perfect with no adjacent bridging continuity. I verified the V2 board had perfect 3.3v 5v GND on the manual install pads coming through the flex. I reflashed the firmware, no change. I re-hot air'd the back of V2 pcb under U1 U2 U4 and had excellent flow through to the chip, still no change. My V2 pcb is an older revision that still had R41 pads and R42 pads, new revision only has R42 and it wants a 1Kohm, my BOM said R41 is 3.9Kohm and R42 1.58Kohm. I noticed on the github Borti uses a blue pcb v2 in the pictures that is actually exactly like my board and he did not have R41 populated. I removed R41 3.9Kohm and no change. I tried zooming in and it looked like the resistor in R42 he had said "308", in the bom R23/R24 use a 308 2Kohm so i tried that with no R41, no change. I verified the board on his github had VCCIO B6 and B7 2.5v jumpers closed for the fpga bank 6/7 and it did, I also had these closed (closed prior to fpga flash also). I microscope'd solder joints and tested every continuity I could trace for many hours.
On U7, U5, U6, and D23 the silk screen has a white dash at a single respective corner, I'm guessing this is the alignment to orientate the dot on the corner of the components. I can verify the clocks at U6/U5 off the github pics follow this logic. I can't tell on D23 or U7 from the github. This is one thing that I cannot verify at the moment. I verified resistance through all resistor arrays. If anybody has any input I would be indebted. The function of the LEDs would be nice to know also. Thank you. (I have a bunch of pics, I just need to figure out how to upload) https://imgur.com/a/IqTQ6LT
Borti N64 Adv. V2 HDMI Troubleshooting
Re: Borti N64 Adv. V2 HDMI Troubleshooting
I had a few of the newest revision pcb's printed up 20231119 and I built out one of those whizoo3 reflow ovens. The new revision does not have U7, R41, or jumpers B6/B7. After reflowing both I checked them with the original flex v20230813 install, both boards worked perfect. I orientated U5, U6, D23 the same way on the new boards so that must not be the problem on the original pcbs. I did flip the orientation of the TVS/ESD diodes on the two new boards as the datasheet states they are unidirectional and not bidirectional. So i then flipped the direction on the original pcb to match and reflowed the hdmi connector. Now it detects an hdmi signal and displays a black screen for five seconds and then drops all signal and redetects the signal again in a continuous loop of detecting and dropping. So at this point I'm not sure what the 202200729(d) problem is. I still never figured out how R41/R42 are to be populated, currently I have the 3.9kohm/1.58kohm populated. After flipping the tvs diodes the green led is now constantly lit and the red led is flashing fast. On the boards that work the green is initially lit and the red flashes, when the hdmi detects a signal the red stops flashing and goes out then when the picture displays the green led goes out and everything works perfect. Any ideas would be appreciated.