We're having issues with a Super Mario Bros table that was pretty much left to rot in a warehouse for who knows how long. Someone had proceeded to work on the sound board at some point in time, but we don't know who, why, when, or if they even were successful— the pin was left as parts when we got to it.

After replacing a couple of the rom chips and the interconnect between the main and auxiliary sound boards, we're at the point that the game works, but we only get the speech cues and no background music. MAMMA MIA.

We the voice callouts from the auxiliary board, but the only audio we ever get from the main board is a single "beep" at boot, and a weak sound at the start of a game. We read that we're supposed to get two beeps, but we don't have any other working Gottlieb pin to confront.

The audio amp works fine, the power supply is fine, voltages on the board are stable and what they're supposed to be, we've done the ground mod just to be on the safe side, there are no obviously damaged traces we could identify, and we tried Marco Albus' test rom (the RAM chips were replaced from whomever worked on the board before us with compatible chips, and they tested fine; the input test also seemed to proceed as it says on the guide, but we don't have an oscilloscope to look into the signals in more detail).

We thought it could be caused by the 65C02P2 cpus, but we swapped them between the two columns, and with others we had, and there was no change (also they tested fine with the test roms). We swapped the AD7528J dac thinking it was an issue with the analog conversion, but it didn't change anything. We swapped the Yamaha 2151 on the auxiliary board with another one from a working pin, and no change (the one we swapped out works fine). We tested trying to feed an analog audio signal to the board output (pin #9 of A6P2, which is connected to the auxiliary board and then mixed into the audio amp) from an external source, and it works fine that way. We've even tried to feed the audio output directly into a separate audio amp, bypassing completely the auxiliary board, but we still got nothing. We've thought it could be the S1 chip (LS161) having issues generating the 2mhz clock from the crystal, but it wasn't (we're grasping at straws here).

Going into the sound test, we get signals on the A6P1 connector, but no sound or tone of any kind comes out of the board. The input test of Marco Albus' test rom shows the cpu recognizing inputs as it is written it should (two blinks for one pin to ground, one blink for multiple, tested all of A6P1).

As far as we can tell, the issue should be with the sound generation within the main audio board itself, but we can't locate where. The absolute last thing we can come up with would be to change every single chip on the board "just because"— which is not really something we look forward to.

"Get a new soundboard" is not the suggestion we're looking for. We want to know what the hell is wrong with THIS one, after all the trouble we went through.

TL;DR: System 3 sound board, all roms are fine. Voice plays, music doesn't. Ran out of ideas, need help.

Thank you so much for your help. We could flash another rom and test the Cpu board too, that'd at least rule another possibility out (or solve the problem).

As for the analog path, we don't think so, we tapped into the output immediately after the DAC with another source, and it works fine (pin 20 of the E2 chip), we can get sound into the amp mixed with the voice without trouble...

edit: we tried the CPU test (all we could try without an oscilloscope, that is), and it seems to be fine as far as we could tell. As I wrote before, we tried the sound test and it does -something- on the A6P1 connector on the sound board, so the CPU should be sending signals over to the sound board itself.

While we were at it, we checked the battery (which was fine), disconnected it to flush the ram to be on the safe side (we already did the factory reset before, without any luck), and reset the game to the factory defaults again. Still no music except for the beep at the start, voices still play just fine.

Thank you for the suggestion! Unfortunately, I don't think we have any issues with the analog part of the sound chain...

If we insert an analog signal (from an external source) at any point of the chain after the DAC on the main sound board, even directly on pin 20 of the DAC itself, we get the audio properly mixed and audible from the speakers of the cab. If we connect a separate (amplified) speaker to the output of the main audio board (bypassing all audio filtering and amplifying of the main and auxiliary board), we still get no music.

What we -do- get from the main board, strange as it is, is:
— a single beep at boot
— a single, cymbal-like sound at the start of a game (I suppose when the music is supposed to start)
— a single, different cymbal-like sound after a while (I suppose when the main music loop ends? I have no idea).

So, we're getting —something— out of the main audio board, but no music.

How is the main audio board in a System 3 supposed to work? Is it only "started" and "stopped" by the CPU (as in, commanded when to begin playing a loop from the rom chips until the current game ends), or is it directly controlled by the main CPU at all times? If it's the former, would there be a way to trigger the start of the audio loop on the bench, perhaps?

If it's the latter, is it possible that the CPU can drive the auxiliary correctly, command the board to start doing something, but then fritz off?

I wish we had another System 3 board to swap in and test, but these things seem to be non-existent here.

Another question: we've pretty much excluded the auxiliary board so far under the assumption that "the one on the top takes care of the voices, the one on the bottom does the music". But on the auxiliary board there's a Yamaha synthesizer— what is its purpose? The chip itself is fine (we've swapped it with one from a Fish Tales too, and they both work), but we get nothing on the output of the YM3014 DAC following it. If we inject an analog signal after the YM3014, we get sound out of that.

Could it be that the DAC on the auxiliary board is toast, and we're not getting any music out of the main sound board because —no music is generated there to begin with—? Again, we have no other comparable Gottlieb system to look at.

Also, the YM2151 is running hot. Not scalding, but unpleasant to the touch.

An update from the pinball that came straight out of hell: we tried swapping out U11 on the cpu board, and things got slightly worse at boot (the game hangs while booting if the ribbon cable to the display board is connected, but if it's unplugged and then reconnected, the game proceeds to boot and work just as before).

The rom on the display board does not match the one we downloaded, but we don't have a spare of that specific chip at hand, so we're not gonna do anything about that. If we get the machine to talk, we'll tackle the intermittent boot issues, otherwise, well...

Thank you for your suggestions! We'll surely do that when we get to it, but for the time being we just put the original U11 back in and the machine boots 100% of the time without issues.
(I personally have no idea why the board wouldn't boot with a different chip than its original one, but again, we'll take care of that when we get to it).

We're trying to get an Arena functional (the closest Gottlieb we actually have on hand), so we'll be able to piggyback on its sound card and test the digital signal into our Super Mario dac, and the digital signal from the Super Mario into the Arena's dac.

Fat lotta good it's gonna do us, but at least the boss'll have a working Arena after that.

New week, new update from our personal hell.
The Arena is functional, it squawks and beeps and bops and all that. We initially thought that the commands sent to the sound board from the CPU in these Gottlieb games were complex, but it seems that (at least on the Arena), it's just pulling one or more lines low and telling the sound board which sound loops to play. Even without a sound test available, we can get it to play the game sounds at will.

The sound test in the Super Mario does just that, it pulls a line low and, supposedly, gets the board to play a simple tone. On the machine we have here, we get the signal low, but no sound.

...which is extremely confusing, given that it still manages to command the speech and sfx just fine. And the starting beep. And the weird cymbal crash sound we hear from time to time.

THIS GAME'S A LOTTA FUN.

Thank you for your offer! Unfortunately we're about an Ocean away, so the logistics of that would be quite complex

If you're willing to tinker (without a soldering iron or anything, just popping out a chip, if you have it on a socket) with your Mario, we're kinda curious to know -what- sounds come out of the main board, and what from the auxiliary. We're kinda doubting the two DACs involved (we've swapped out the one on the main sound board, but we don't have a spare for the one in the auxiliary board), and some things are hard to tell from youtube videos and emulated tables.

What music plays/what is missing if you remove the Yamaha 2151 (U9) from the auxiliary board?

What music plays/what is missing if you remove the D-rom chip from the main sound board?

We are definitely sure that the cymbal sounds are coming from the main sound board because we've isolated the audio output from that (Pin 9 of A6P2) and connected to a separate amplified speaker, and we get those three sounds (one beep at boot, a cymbal at the start of a game, and another cymbal after a while during an ongoing game) and nothing else...

We've tried comparing the audio boards from the Arena and the Super Mario, but it seems that, even though most of the board layout and components are the same, they process the audio inputs from the CPU in a different way:
— when one of the S0-S7 inputs of the board gets pulled low (even shorted directly to ground), the Arena board plays some of the game sounds. Pin 8 of A2, which is normally low, briefly changes state (enough to tell with a multimeter), and then goes back to low.

— on the Super Mario, pin 8 of A2 is low at boot. Whenever a sound plays (not the boot sound, but a sound sent from the CPU, like starting a game or inserting a coin), the pin goes high and stays high. If we pull one of the S0-S7 lines low (either via the sound test, or manually), no sound plays. If we manually pull pin 8 of A2 low, we get the board to repeat the last sound played (any of the speech samples), and it goes back to staying high.

— if we disconnect the A6P1 connector on the Super Mario (which carries the S0-S7 from the CPU board), the A2 on the sound board behaves like the one in the Arena— pin 8 is normally low, changes state when an S line is pulled low, and goes back to low. Still no sound.

We're comparing two different generations of the same system— maybe the System 3 one has stricter timings for the inputs, so the difference is by design. Or maybe there's something amiss with the via at U5 on the Cpu board— but it can play the sound clips when it's supposed to, so...

(Also— apparently we're supposed to hear the coin sound when the ball hits the mushroom bumpers, but we don't, so we might not even be getting all the sounds we should...)

Thank you! We'll try that out and see if we can identify what comes out of where in the game (we've tried the procedure to boot into the other sound test, but it doesn't seem to work).

In the meantime, we've asked a colleague for a favor and we're having him put the two 6522 vias on the CPU board onto sockets, so we'll try to swap them out and see if, by any chance, we're just getting funky signals from the CPU board on the audio lines (unlikely, but we've ran out of ideas about twenty tries ago).