Hey guys. This is my first post. I apologize if I made any mistakes.
I recently bought a famiclone on aliexpress with PPU FPGA and RGB output called RGB-FC. Unfortunately I made a mistake when I put a 150 in 1 multicard that had a short circuit on its PCB (my stupidity), when I turned on the famiclone the led went off, indicating a short circuit and when I removed the cartridge and turned it on again it didn't work anymore, it turned on only the led with a gray screen, as if it did not recognize the CPU or PPU FPGA. How can I tell if the problem is in the FPGA? Is there any way to test the FPGA with just a multimeter? Would a famicom start a game, even with artifacts, if damage occurred to the vram or buss buffer (74HC368AP)? Searching the fpga works with 3V and the famiclone with 5v in the slot. Another thing, I can change the output resolution (480p or 1280x1024) in addition to enabling and disabling scanlines. Is that a good sign that the back of the slot is ok? So the problem could be in VRAM, BUSS buffer or CPU?
[img][ https://imgur.com/a/Jl7fQgu]
[img][ https://imgur.com/a/Kv1racW]
RGB FC Aliexpress
-
NewSchoolBoxer
- Posts: 369
- Joined: Fri Jun 21, 2019 2:53 pm
- Location: Atlanta, GA
Re: RGB FC Aliexpress
I can't give you a very good answer but no response so I'll try. The 150 in 1 multicarts are among the worst flash carts ever made and can and have destroyed real NESes after several months of use due to negligent 5V <-> 3.3V conversion. Now probably the famiclone runs off 3.3V and translates to/from 5V from carts correctly. If cart is sending 3.3V directly and famiclone still works...I think is okay but I haven't examined the circuits in full.
Second pic is blank. First pic isn't zoomed in enough so hard to tell what's going on. Is the U6 the voltage regulator? Is there a fuse or a diode for reverse voltage protection? You can do some basic troubleshooting with the multimeter continuity test, which is the diode setting. While console turned off in case that isn't clear. If you get a beep with one lead on one part of the circuit and other lead on the other, has like 50 ohms or less resistance and connection passes DC just fine. As in, the fuse still works, the traces are intact, can also check diode/LED and which way the orientation is because meter will only beep one way. Doesn't work testing continuity with capacitors since they block DC.
You'd really want to know which pin(s) had the short circuit and then lookup what the pin number(s) connect to on a normal NES and then compare on the famiclone. Since that's what got fried. You can't really test the FPGA or VRAM with a multimeter. You'd need a digital logic analyzer. That said, if you fried the circuity from the pin, I'd assume you would always measure 0V versus 3V when logic is supposed to be high. You could measure the AC voltage even if not true RMS. If > 0V then it's maybe okay.
Here is the first NES cart pinout I found. You see darn near everything connects directly to the CPU or PPU (so the FPGA?), 2 connect to VRAM, 2 to audio, 1 to CPU clock, 1 to system/master clock for 72 pin cart support. Very nice that it shows the direction the data goes. Damaged VRAM can still allow NES to boot but with garbled graphics. Is replaceable. Worth doing if you suspect damage from pin mapping.
74HC368AP has ESD protection but not short circuit, at least for the one datasheet I looked at. Frequency would be in the kHz or MHz range so checking the inverter is really working and inverting would only be feasible with DC. Same AC voltage measuring on the output means it's maybe okay. You could de-solder it to properly test. Internet says the CPU's R/W line as well as the audio amplifier are connected to the hex inverter. Frying a CPU pin could chain into frying the inverter but not guaranteed.
Second pic is blank. First pic isn't zoomed in enough so hard to tell what's going on. Is the U6 the voltage regulator? Is there a fuse or a diode for reverse voltage protection? You can do some basic troubleshooting with the multimeter continuity test, which is the diode setting. While console turned off in case that isn't clear. If you get a beep with one lead on one part of the circuit and other lead on the other, has like 50 ohms or less resistance and connection passes DC just fine. As in, the fuse still works, the traces are intact, can also check diode/LED and which way the orientation is because meter will only beep one way. Doesn't work testing continuity with capacitors since they block DC.
You'd really want to know which pin(s) had the short circuit and then lookup what the pin number(s) connect to on a normal NES and then compare on the famiclone. Since that's what got fried. You can't really test the FPGA or VRAM with a multimeter. You'd need a digital logic analyzer. That said, if you fried the circuity from the pin, I'd assume you would always measure 0V versus 3V when logic is supposed to be high. You could measure the AC voltage even if not true RMS. If > 0V then it's maybe okay.
Here is the first NES cart pinout I found. You see darn near everything connects directly to the CPU or PPU (so the FPGA?), 2 connect to VRAM, 2 to audio, 1 to CPU clock, 1 to system/master clock for 72 pin cart support. Very nice that it shows the direction the data goes. Damaged VRAM can still allow NES to boot but with garbled graphics. Is replaceable. Worth doing if you suspect damage from pin mapping.
74HC368AP has ESD protection but not short circuit, at least for the one datasheet I looked at. Frequency would be in the kHz or MHz range so checking the inverter is really working and inverting would only be feasible with DC. Same AC voltage measuring on the output means it's maybe okay. You could de-solder it to properly test. Internet says the CPU's R/W line as well as the audio amplifier are connected to the hex inverter. Frying a CPU pin could chain into frying the inverter but not guaranteed.
-
casuais_play
- Posts: 6
- Joined: Sun May 07, 2023 8:43 pm
Re: RGB FC Aliexpress
thank you for the answer
I managed to fix this unit, I replaced the 74HC368AP, in fact I was having a bad cart and everything is resolved. But I had another unit with a strange problem. This famiclone uses an original CPU (RP2A03G) and a PPU RGB FPGA with video output in RGBs and works with a 5V via a USB-C input https://imgur.com/1g3ytle . Recently I noticed strange behavior with few games and 240p suit when using with N8 clone. For example: In 240p suit, graphics corruption occurs on the SMPTE test screen after a few seconds, depending on power supplie I am using, this effect takes time to occur https://imgur.com/lsNK4NS . In Super Mario Bros 3 the game crashes or starts automatically after a certain period on the title screen https://imgur.com/PWW7nRD . In games like 1943 or Little Nemo, I experienced crashes. I tested on other famiclones and didn't get the same results, ruling out the N8 as being the problem. Does anyone have any idea what could be happening? I tested with different power supplies and the behavior is more noticeable or attenuated. Could it be a noise problem caused by the switching power supply?
I managed to fix this unit, I replaced the 74HC368AP, in fact I was having a bad cart and everything is resolved. But I had another unit with a strange problem. This famiclone uses an original CPU (RP2A03G) and a PPU RGB FPGA with video output in RGBs and works with a 5V via a USB-C input https://imgur.com/1g3ytle . Recently I noticed strange behavior with few games and 240p suit when using with N8 clone. For example: In 240p suit, graphics corruption occurs on the SMPTE test screen after a few seconds, depending on power supplie I am using, this effect takes time to occur https://imgur.com/lsNK4NS . In Super Mario Bros 3 the game crashes or starts automatically after a certain period on the title screen https://imgur.com/PWW7nRD . In games like 1943 or Little Nemo, I experienced crashes. I tested on other famiclones and didn't get the same results, ruling out the N8 as being the problem. Does anyone have any idea what could be happening? I tested with different power supplies and the behavior is more noticeable or attenuated. Could it be a noise problem caused by the switching power supply?