Super Space Invaders '91 - Repair thread

[bobmoo79]

I've pulled the EPROMS and verified IC8, IC9, IC38 that I programmed previously and they checked out ok against the MAME romset I originally used (SSIb) and a newly downloaded version (SSI).
I also pulled IC1 (C64-02) and it verified ok too (using Mitsubishi M5L27512K as the IC specified in my programmer).

That only leaves IC51 to check but unfortunately it's a 44pin IC and my programmer only has a 42pin socket so I'll need to buy an adapter for that one. My plan is to replace it with a 27c800 if I see any issues with the original ROM.
In the meantime I'll see if I can check out the address and data lines at the 68000 and on a rom and check the control lines going to the main ROMs, but it'll be a few days before I can get to it.



For the ROM programming, I'm not doing anything clever - just loading the ROM file into my programmer software and then programming the ROM. I'm not messing with any settings. Hopefully that should work, but if you're aware of anything that could go wrong with that, please let me know?

 

[Azure Oz2]

To get the game to a minimal running mode you should only need IC8 and IC9.

IC38 will be for Z80 sound code and should not prevent the 68000 from starting. If it does control startup of the 68000 there will be no activity on the 68000 until it starts it, then it should run normally. I would assume it doesn't until testing tells us otherwise.

Any other ROMs are most likely, graphics or sound related and not important to getting the 68000 running, so I would not be spending time on them now. You can deal with those later if they prove to be faulty.

For IC8 and IC9 if you can verify them against the files you loaded they should be correct. Make sure they are in the correct sockets, if they are swapped the 68000 will be reading the instructions incorrectly. You can them swapping them to see if it makes a difference, can't work any less than it is. Then swap them back if you believe they were correct in the first place.

Then next thing would be to check the ROM sockets, they look like there might be corrosion on a few pins that maybe causing bad connections. If you can try and find where the data and address lines go from the ROMs and from the 68000 you should be able to locate if there are any buffers involved or if they are direct wired and check continuity between the data and address pins.

Hopefully there is a something there that fixes things. If not the next step would be to check the !CS and !OE pins on IC8 and IC9 to see which are tied active (low) and which are pulsing (active low) after reset before it stops.

 

[bobmoo79]

Ok some great tips there that I'll definitely look into. Will just be a few days before I'm able to get back to it though because of other things going on. Thanks for all your help so far, it's been great. I'm learning a lot that's for sure

 

[bobmoo79]

27c2001 pinout

 

[bobmoo79]

Although I know I didn't need to, I checked IC51 against the MAME ROM file using the adapter I bought and it's fine. So we know all ROMS are good.



Yes these ROM (IC8 and 9) sockets are corroded and causing connection issues, even when trying to buzz out these traces so I will need to clean/replace these. I was hoping to avoid it but I don't think that's an option now.

I'm still tracing out connections and the Address lines are direct wired (no buffers). One notable discovery was that A14 of IC9 was not connected to the same pin on IC8 or the equivalent on the SRAM ICs or CPU. See below, taken from the Final Blow schematic.



I corrected this and tested but it didn't make any difference to the behaviour on !AS, !DTACK, !HALT, or !RESET.

I've also been checking all traces from IC3 (CPU) to:

IC2 = TC0530SYC (Sound comm chip)

IC4 = uPD43256 (SRAM)

IC5 = uPD43256 (SRAM)

IC8 = 27c2001 (ROM)

IC9 = 27c2001 (ROM)

IC25 = TC0540OBN (object chip)

IC48 = TC0260DAR (Custom chip - used for RGB output?)

IC10 = TC0520TBC (object chip)

IC54 = TC0220IOC (Reset and IO chip).

Another point to note is for the uPD43256 SRAM ICs (IC4 and IC5), the !CS are tied together and the !E are also both tied together (see photo - orange bodge wires installed before I got the machine). These being joined aligns with the schematic above.

Note: the blue bodge wires in the photo were installed by me to correct torn traces that I could see needed fixing before my first power up. I'll tidy those up if I ever get this working).

 

[bobmoo79]

The schematics for Final Blow show many similarities to this PCB and so I have been using it to help navigate the board, checking connections etc.

https://arcarc.xmission.com/PDF_Arcade_Manuals_and_Schematics/Final%20Blow%20Schematics.pdf

 

[bobmoo79]

I grabbed this capture of!AS against !RESET as well. I don't think it's enlightening but I didn't post it previously

 

[bobmoo79]

Broken trace found: CPU Pin 33 (A4) not connected to ROM (pin 6) / RAM (pin 8).

The attached photo shows what is under ic4. Somebody has used an abrasive, removing the solder mask from multiple traces and damaging some. Only one is broken but I might run patch wires for one of two others to make sure they don't fail in the near future.




Also, another broken trace found: IC9 (ROM) PIN21 to IC4 (RAM) PIN19. This is for I/O8.

 

[bobmoo79]

I repaired the broken traces and still no good.
No change to the behaviour at all.