What's new

CPS3 no boot - UltraBIOS cart, screen has garbled pattern

SheepyChris

SheepyChris

Student
Joined
Oct 3, 2024
Messages
37
Reaction score
29
Location
Canada
Hey y'all,

We've gotten around to start work on a CPS3 with a SFIII Third Strike cartridge at our arcade, the battery on the cart ran dry and the screen would either display nothing or a garbled pattern typical of that. We've removed the battery, desoldered the flash chip, installed Darksoft's UltraBIOS custom SH2 with CD, resoldered and there was no change to its behavior. We've done the following so far:
- Double checked orientation and solder on flash chip
- Reseated the cartridge and SIMMs several times alongside using isopropyl alcohol and contact cleaner on pins
- Reflowed and continuity tested every pin on the cartridge connector
- Reflowed pins on the GLL1, CCU and M514260B chips that looked tarnished/corroded
- All visible traces that looked suspicious of corrosion were tested, including resistor pack traces between CCU and M514260B
- Checked voltages, test cabinet was delivering 4.92v

The only things I haven't gone through yet are:
- Testing all vias near the cartridge port
- Replacing the 6.25MHz crystal oscillator on the cartridge - it has some severe surface corrosion as seen in the pictures
- Test the cart on another PCB (don't have access to a second one :p)

If there's anything we're missing any help would be much appreciated.
 

Attachments

  • 20251015_002016.jpg
    20251015_002016.jpg
    310.7 KB · Views: 52
  • 20251015_002026.jpg
    20251015_002026.jpg
    298.6 KB · Views: 54
  • 20251014_225514.jpg
    20251014_225514.jpg
    297 KB · Views: 48
PolloLoco said:
@SheepyChris if you are based in the Vancouver/Lower Mainland area, I have a working CPS3 board I can lend to help test the cart.
Click to expand...
I'm all the way in Ontario so wouldn't work out lol.

Currently back at it again, testing things with a new PSU so will update on progress here shortly

Update: new PSU with rock solid 5.01v had no change, also went through all visible vias near the cartridge slot and all tested good. Time to focus on the cart itself perhaps?
 
Last edited:
Another update, my friend spotted a pin on the cart flash chip not connected and reflowed all of the pins. We now get a black screen when starting the system up, progress I guess? We flashed the UltraBIOS for custom SH2 CD version as we do have the CD drive, although it's not clear to me if the normal SH2 version will work on this cartridge as well. If there's any guides to properly flashing the software please let me know as every bit of information I've found is scattered across so many forum posts and most of what has been done essentially became a guessing game.
 
A couple of questions. Did the battery leak? Even if you cleaned, if it leaked, you should look for damaged traces.
The Crystal on that cart looks super corroded, might not work well. The custom SH2 could also have some corrossions. This needs to be checked.

The BIOS that you need is the CUSTOM, so you are good on that one.
 
Also: recheck your solder work. Then check it again. If the cart isn’t damaged & if the bios chip is properly flashed w/all pins connected you *will* see it boot.
 
The original battery was in pristine shape and without any leaks, only the crystal has corrosion on its shell (maybe due to humidity?). We pulled out the flash chip and reflashed it with the custom CD binary again as there was one byte that tested different (FF vs 80), currently reinstalling it on the cart. Would the CD drive also need to be connected on its first boot?
 

Attachments

  • 20251014_191751.jpg
    20251014_191751.jpg
    263.7 KB · Views: 34
  • 20251014_191800.jpg
    20251014_191800.jpg
    220.7 KB · Views: 33
I would also clean the corrosion off the lower part of the pins. It looks like one of the pins just to the right of the notch in the cart may not be making good contact.
 
Corrosion has been cleaned off the pins that still had a few remnants of it, no change and it still remains at a black screen. Went through all pins on the flash chip and CPU, they are correctly soldered and testing GOOD being probed up to the end of traces. I need to bring my oscilloscope to the arcade one day and see if that oscillator is indeed working as I couldn't find any suitable replacements online for 6.25MHz (although I've seen repros use 6.144, how tolerant is the system to different clocks?) 😬 The main board crystal looks equally crusty but seems to be operating given there's still garbled video output without a cartridge plugged in. Does anyone in Ontario have a working CPS3 and is able to lend a hand with diagnosing if it's the cart or motherboard at fault here?

On a side note I've noticed the cart's CPU is labeled as D, and there's a thread here talking about how some have been unsuccessful in getting certain revisions to work, would this have anything to do with what I'm facing here? At this point we are willing to buy another working cart either converted or repro'd.
 
Last edited:
Necroing this thread because I'm continuing onwards with replacing the crystal (oscillator?) on the cartridge soon but have noticed a big discrepancy on what type of replacements are used - the original 6.25MHz frequency is nearly unobtainable outside of through-hole crystals from Parallax, meanwhile I've seen 6.144MHz and even 6MHz frequencies being used in repro cartridges. Is this crystal utilized by the SH-2 CPU and if so, could lower speeds introduce slowdowns and negatively affect gameplay in 3S? Can higher frequencies be used to combat this theoretical scenario? Could this become an existential crisis to FGC tournament organizers? Half joking there but I seriously couldn't find any information about this and don't want to commit to spending and blindly throwing more parts at this.
 
You must log in or register to reply here.
Back
Top