Solved - Sinistar Freezes During Title Screen and Gameplay

[SNESNESCUBE64]

Here's what I tried this evening: Replaced the sockets for the decoders and CPU, those were a bit yucky. No change. Burnt a new ROM set for the off chance something goofing around. No change. I am going to start poking around tomorrow, I'm gonna start by poking the 153s, I've had terrible luck with those. Any other ideas of what I can do?

 

[SNESNESCUBE64]

Managed to get this exactly once. Odd that the background was yellow for this. Sometimes if I run the RAM test, through the coin door buttons it will say ram 21 or 22 (they are both new), once it said 24.

 

[mecha]

Managed to get this exactly once. Odd that the background was yellow for this. Sometimes if I run the RAM test, through the coin door buttons it will say ram 21 or 22 (they are both new), once it said 24.
View attachment 862644

don't bother with 74153s. that's going to do with address bus, which the game runs, so it's not there.

your ram error 21 or 22 could suggest something may be amiss with 3H 74373 (assuming this is an original Rev. D MPU; otherwise it will be a 74374 part)

logic probing inside the game will be kind of nightmarish, but I would compare data in/out at the 21 and 22 rams against 31 or 32. particularly when the game craps out. I'm not really a rocket scientist like braedel, I just play one on TV, without having it in front of me I can't really say. I'm also fighting narcolepsy right now... we'll revisit this later.

 

[SNESNESCUBE64]

your ram error 21 or 22 could suggest something may be amiss with 3H 74373 (assuming this is an original Rev. D MPU; otherwise it will be a 74374 part)

That could make sense. I didn't consider it because I figured I'd have more problems being it is with the dram. But I suppose the occasional miss with latching would cause problems. I'll take a peek after work today. Thanks.

 

[mecha]

That could make sense. I didn't consider it because I figured I'd have more problems being it is with the dram. But I suppose the occasional miss with latching would cause problems. I'll take a peek after work today. Thanks.

is there any battery leak anywhere? it kind of wrecks things in that area.

 

[SNESNESCUBE64]

is there any battery leak anywhere? it kind of wrecks things in that area.

Nothing serious. Little bit of green on the holder solder points and the holder was trashed.

I made myself a remote battery pack to keep em off the board from now on.

 

[SNESNESCUBE64]

your ram error 21 or 22 could suggest something may be amiss with 3H 74373 (assuming this is an original Rev. D MPU; otherwise it will be a 74374 part)

I think you nailed it. I piggybacked one and it ran a whole hour without crashing. As soon as I pulled it, next time it went into gameplay it crashed. I went ahead and swapped the 74ls373 at 3H and am gonna let it run for a while. I also put a fun message on the title screen, it's neat that I can do this. It passes the tests on my chip tester but I cannot argue with results. It must fail under stress (hey me too).

 

[vintagegamer]

Rock. Awn.

Gorgar's second cousin??

 

[SNESNESCUBE64]

Gorgar's second cousin??

The Gorgar pinball used the same sound and speech board. I figured it would be funny to put that because they are both hungry.

 

[mecha]

The Gorgar pinball used the same sound and speech board. I figured it would be funny to put that because they are both hungry.

Gorgar ftw

I need to reinstate "mecha was here" on Joust and Sinistar

 

[ArcadeTechGW]

Yikes!
View attachment 862509

So, the only way the the screen image would persist(frozen) is if the CPU was halted or not running. Whatever was left in the memory when code execution stopped would continue to be rendered on the display.

~Brad

Where did THAT come from?

 

[mecha]

Where did THAT come from?

I've seen that before now that I remember. there was a period in 2020-2021 that a lot of my pictures weren't saved though, so I might not have evidence.

 

[ArcadeTechGW]

I've seen that before now that I remember. there was a period in 2020-2021 that a lot of my pictures weren't saved though, so I might not have evidence.

I understand that. I'm trying to understand what caused that effect.

It doesn't look like rust to me. It looks more like something took a dump in the area.

 

[SNESNESCUBE64]

It looks more like something took a dump in the area.

I sure hope not. I have no idea what it was. One thing I noticed on this board is that there are a couple spots with corrosion. Not sure if it was bugs or mice or whatever. I spent a lot of time cleaning this cabinet...

I've cleaned it off since there hasn't really been anything severe. This Sinistar seems to be someone's deconversion project that was ended up being sold. For example, it is missing the PCB mounting plate. I went ahead and designed a new plate that I am gonna have made (with extra studs for my remote battery board).

 

[ArcadeTechGW]

That brown material looks biological more than corrosion.

Why not rust? Well, no steel in this area. The solder mask is tin, the legs are tin, so no place for rust.

Now look at the pattern - it almost looks like mold or some bacterial form, rather than corrosion.

 

[mecha]

That brown material looks biological more than corrosion.

Why not rust? Well, no steel in this area. The solder mask is tin, the legs are tin, so no place for rust.

Now look at the pattern - it almost looks like mold or some bacterial form, rather than corrosion.

you're be amazed at the things you find in barn stored games. like almost every one I worked on in Texas had some kind of chemicals and contaminants and overall shit and rust and whatever in them.

 

[SNESNESCUBE64]

you're be amazed at the things you find in barn stored games. like almost every one I worked on in Texas had some kind of chemicals and contaminants and overall shit and rust and whatever in them.

The bottom of the machine was covered in dirt. I still have to do a better job cleaning this but I was itching to get it working.

 

[ArcadeTechGW]

you're be amazed at the things you find in barn stored games. like almost every one I worked on in Texas had some kind of chemicals and contaminants and overall shit and rust and whatever in them.

I likely would be amazed.

I've learned that the best way to figure out something is to sample it and get it tested. Of course, that was when I worked for a company who made billions of dollars a year, and when something was a problem, you darn well better be able to tell the manager what the actual problem was.

Arcade games are in more of a survival/salvage state, so no big budgets to use Scanning Electron Microscopes and Gas Chromatographs to figure out what stuff is.

I'm simply providing my thoughts.

 

[tstone]

In my experience, biological fouling typically has more of a sour/bitter/ammonia taste and a smoother texture.

Rust fouling has more of sweet/tin taste and more of a gritty texture.

 

[SNESNESCUBE64]

One thing I noticed is that SOMETIMES on a cold start, meaning it hasn't been on for several hours, it will throw a random RAM error code (I've seen 13, 21, and 37). It will boot after displaying the error code and crash shortly after (presumably because RAM didn't get properly initialized). Game works fine after either a manual reset or it crashes.

I'm thinking I have one of two problems: something in the reset circuit has shifted a bit out of tolerance or it is taking longer for one of the voltage rails to get up to spec. I see that it performs POR based on both 12V and 5V. The power supply has been rebuilt so I am wondering if maybe that 22uF cap in the reset circuit has drifted a bit or something.