Sega Genesis controller weirdness

[GiGaBiTe]

I rescued a Sega Genesis model 1 (either a VA5 or VA6) that was in pretty hideous condition from the local junk yard. The front of the console had sat in a pool of liquid at some point and turned into a rusting hulk on the inside. After about 5 hours of rust removal, trace repair, recapping and replacing two blown 7805s, the console lives again.

Though it soon developed a weird fault where the second controller port will randomly generate button presses after a time. At first I thought it was the port itself, so I removed it. After running the console a bit, the random inputs kept happening. So I went up the food chain and removed a line of these unlabelled 3 pin SIP chips, which I think are some sort of inductors and the problem went away, though it also made the up/down buttons on the first controllers D-PAD stop working.

These traces are already in pretty poor condition, I don't want to play musical chairs with inductors that I can't tell are bad or not. Is there a replacement for these parts?

[GiGaBiTe]

Well I managed to fix the second controller acting weird, but another issue popped up with the sound, it cuts in and out randomly and has crackles and popping.

I replaced all of the capacitors on that side of the board and it's still happening, leading me to believe it's a chip issue.

[Deleted]

Good work on the restoration so far!

I'm not sure what would be causing the issue with the sound, but if it started happening after you fixed the controller 2 weirdness, I'd see about undoing some of that to see if it goes away.

[fisher]

Nice restoration work!! Congratulations!!
I had a similar problem some time ago.
It was the power switch that was making the sound to get lots of noise.
You can confirm it by slowly moving or gently beating on the power switch and see if there's any difference in the sound.
Good Luck!!

[GiGaBiTe]

I managed to fix the audio problem, it was one of the two blue caps on the top right. It had high ESR and started pulling heavy current and overloading one of the LM7805s.

Replacing the rest of the caps made the board behave properly again.

Now I just need to get some S-Video connectors and make up a board so I can get good video output.

[wraith]

Rise From Your Grave !!!

[GiGaBiTe]

Did a bit more work on the console in preparation of overclocking it:



I added a socket so I could test a CPU that I got an electronics lot some years ago. It's an engineering sample of a 68000 rated at 16 MHz (EFX68000C16) made in week 22 of 1985. I also got three oscillators and a socket so I can test 10, 12 and 16 MHz clocks. I don't expect 16 MHz is going to work, but the oscillators were $2 a piece so it would be an interesting experiment.

I'm sort of confused what the 220 ohm resistor is needed for on pin 6 (address strobe) and why it's not directly connected, but I replicated it from the original CPU to make sure it was in the same state for no problems.

[GiGaBiTe]

Its been awhile since I last worked on this unit, but I'm in the process of designing an "A/V Module" which replaces the RF modulator:




It provides RCA, S-Video and two switches for JP/US and 50/60Hz. An additional yet to be design part will screw into the upper front portion of the module to provide two more switches for overclocking support (68000 halt and clock selection switch.)

I will eventually release the files for the pieces of the entire module, but the user will be required to source the components.

[GiGaBiTe]

So I finally wired up a socket for the clock oscillator on a piece of protoboard. The initial test resulted in a dead console with the occasional garbage on the screen. After reading a bit about clock signals online, I found out I made the clock line far too long. In addition, I wasn't using shielded wire, so the signal was probably complete garbage going into the 68000.

I replaced the clock wire with a shielded grounded wire and made up a RF shield for the back of the proto board out of aluminum foil and the console now started working at 10 MHz. The sound gets a bit funky at times, likely because the CPU and system clock are out of sync. Both 12 and 16 MHz don't work at all. 12 MHz results in lots of graphical weirdness and frequent crashing, but it will sometimes stay stable if I hold my hand on the aluminum RF shield, leading me to believe there's still a signal degradation issue. 16 MHz will rarely work at all, but when it does, there's usually heavy graphical corruption and the sound doesn't work at all.

I'll probably have to design a PCB with a proper ground plane and bypass capacitors to solve the signal degradation, and make a proper RF shield can instead of an aluminum foil sheet.