Williams Stargate reset

72gto

Well-known member

Donor 6 years: 2017, 2021-2025
Joined
Aug 14, 2017
Messages
1,103
Reaction score
393
Location
Ohio
I have a new Stargate that is having an issue.

At power up the game boots and does a system check and gives the message all systems go and the segmented led reads 0. I have normal start up sounds.

However the game doesn't start and gives the message "high score table reset adjustment failure." In other posts this was mentioned in the context of the battery upgrade and how the high scores are stored. However this PCB has a lithium battery mod and I replaced the battery while troubleshooting.

The following may be related, the game seems to use 12v power from the linear power supply
But 5v from a switcher. Williams factory design uses ac voltage for the coin door and this looks hacked to hell because it's using doc voltage from the switcher now. I could use some verification that the test switch is wired correctly to the interface (?) board. I have a feeling that is causing another issue that apart from getting stuck at the table I can't manually reset to game over mode to fix it.

Any suggestions-- I can't advance into any menu auto or manual I can only reset and get the same error.

Thanks everybody!
 
I've done the hybrid power supply thing a few times this year, and 2 of them at work I wound up undoing because they progressively chipped away at the CMOS. I was always under the impression that it was unregulated +12 that handles that CMOS protection thing, which may be true, but the +5 bleeds off instantly when you shut the game off and that's half of how the system works, it's supposed to steadily reduce the +5 to the CPU on power down.

you most likely have a bad CMOS, which is an incredibly common thing. if the linear power supply is still in there, I don't understand why they ran the coin door lights to the switcher, that plug rarely has a potential to burn up, and the ceramic resistor inline there (if it's present) can have a meltdown too.

you should be able to press the Advance button on the test panel to skip the high score table message -- I've found a couple boards as of late that throw the "table top" error that I could bypass with Advance too. but with the straight switcher voodoo, many people neglect to port the coin door ground wire from the GI plug mentioned above (as it's not part of the same chain as the controls) to the ground on the switcher. it's a black-orange wire, I'm going to lean towards it being cut off the old plug and just hanging inside somewhere.

if you can get it to a point where you can Advance past the error, take a look at the high scores. if they're like all 777777, and you go into settings mode and every option is like 777777, the CMOS is definitely hosed. you can attempt to manually factory restore, reset high score table and reset bookkeeping, but the 7's usually mean you can't fix it without replacing that chip. which conveniently is soldered in. :)

I'll PM you my phone number, it's easier to text me if you need additional help.
 
That explanation helps clarify the corruption occurring. Right now I can't manually reset. But l see loose wires coming off of the test switch advance so there's probably no signal going to the pcb to move it in the menu even apart from the ground issue. I was going to use my Joust machine to try to identify which wires should be going into the respective plugs of that molex (from the coin door immediately below the control panel) so I could at least figure out the correct pin and get signal to it to manually move to game over mode. I also had two other ram boards on hand to test so I got a little different response with the second board that I may have to troubleshoot separately but it looks to be generally the same description with bookkeeping so that appears to confirm that it's a power issue.

(Also I got your PM and sent you one back--thanks buddy!)
 
I've done the hybrid power supply thing a few times this year, and 2 of them at work I wound up undoing because they progressively chipped away at the CMOS. I was always under the impression that it was unregulated +12 that handles that CMOS protection thing, which may be true, but the +5 bleeds off instantly when you shut the game off and that's half of how the system works, it's supposed to steadily reduce the +5 to the CPU on power down.

you most likely have a bad CMOS, which is an incredibly common thing. if the linear power supply is still in there, I don't understand why they ran the coin door lights to the switcher, that plug rarely has a potential to burn up, and the ceramic resistor inline there (if it's present) can have a meltdown too.

you should be able to press the Advance button on the test panel to skip the high score table message -- I've found a couple boards as of late that throw the "table top" error that I could bypass with Advance too. but with the straight switcher voodoo, many people neglect to port the coin door ground wire from the GI plug mentioned above (as it's not part of the same chain as the controls) to the ground on the switcher. it's a black-orange wire, I'm going to lean towards it being cut off the old plug and just hanging inside somewhere.

if you can get it to a point where you can Advance past the error, take a look at the high scores. if they're like all 777777, and you go into settings mode and every option is like 777777, the CMOS is definitely hosed. you can attempt to manually factory restore, reset high score table and reset bookkeeping, but the 7's usually mean you can't fix it without replacing that chip. which conveniently is soldered in. :)

I'll PM you my phone number, it's easier to text me if you need additional help.

Hey Mecha happy easter. Older thread but my Stargate has me scratching my head and you got me wondering. I am using a switcher, I do have rebuilt linear. You think it's possible to put a cap or something on the +5v line to hold it up when powered off a few. Sometimes random scores, high score will be gone and it will move up other scores. They are not all erased just randomly removing some, not from the bottom. Never got these to work w a switcher?
 
braedel actually came up with a zener diode mod to keep switchers from wiping the CMOS chips. all that stuff is so far out of sight and out of mind to me right now I don't remember where. there are a lot of CMOSes that are just bad too though. pay attention to the score table, if it's drawing random text all over the place and you have like question marks and other goofy symbols showing, that's a bad CMOS. also if the scores blank out with 4000 or 10,000 point scores in chunks over time or if it resets without a power cycle, bad CMOS.
 
braedel actually came up with a zener diode mod to keep switchers from wiping the CMOS chips. all that stuff is so far out of sight and out of mind to me right now I don't remember where. there are a lot of CMOSes that are just bad too though. pay attention to the score table, if it's drawing random text all over the place and you have like question marks and other goofy symbols showing, that's a bad CMOS. also if the scores blank out with 4000 or 10,000 point scores in chunks over time or if it resets without a power cycle, bad CMOS.

Definitely the later, 4000 point scores chunks missing. Thank you, you say that CMOS is soldered on huh. Time to investigate :D
 
yeah that's the one part that kind of sucks but if you have good tools and technique it's not the end of the universe. it's preferred to socket the new one, but I ran out of sockets at a point and started hardwiring the new ones in. I'm a proponent for the 2032 lithium battery mod, anti-NVRAM, if you get corrupt CMOS you can't interrupt the power supply to blank it back. that's why having multiple different game rom boards on hand is useful, you can eventually get the game to recognize something's wrong and force its own factory restore.
 
yeah that's the one part that kind of sucks but if you have good tools and technique it's not the end of the universe. it's preferred to socket the new one, but I ran out of sockets at a point and started hardwiring the new ones in. I'm a proponent for the 2032 lithium battery mod, anti-NVRAM, if you get corrupt CMOS you can't interrupt the power supply to blank it back. that's why having multiple different game rom boards on hand is useful, you can eventually get the game to recognize something's wrong and force its own factory restore.

Nice! Definitely looking into. Found this interesting read too. They put a 47000-18000 uf cap on the +5. Will try and test that then install a socketed cmos

 
Back
Top Bottom