(Topic ID: 294786)

Gottlieb system 80 sound/speech board - I need help!

By amnesia

4 months ago



Topic Stats

  • 9 posts
  • 5 Pinsiders participating
  • Latest reply 3 months ago by G-P-E
  • Topic is favorited by 2 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

20210608_065405 (resized).jpg
20210608_065402 (resized).jpg

#1 4 months ago

HI all,
I need help from someone who knows way more than I do about electonics!
I'm trying to troubleshoot an A6 sound/speech board from my Black Hole.
In a nutshell, the problem I'm experiencing is that when I power up the A6 Sound/Speech board it immediately makes the first tone of the A6 "test", and no further sound - despite the game working perfectly in silence. I figured out that if I press the sound board test button, it makes another tone, and if I press it again it says "Complete" and the sound works as normal. It's as if pin 17 of the 6532 RIOT is grounding somehow during boot up and thereby innitiating a "test" routine. After much experimentation and probing I have figured out all the following:

1) all input lines are working perfectly from the CPU board Z27 > Z31 > driver board Z13 > sound/speech board U16 & U17 > 6532 RIOT

2) after the boot failure if the machine is powered off, and then powered back on within a few seconds to a couple of minutes, it powers up normally - without going into the "test" routine. Once running, all sounds are present, and the machine works perfectly. As well, all tests from Tony Holdgate, Clay, & Kirb's trouble shooting guides check out 100% ok!

3) When the board fails to boot, the RESET lines of the 6502 (pin 40) and the 6532 (pin 34) both go Lo briefly then Hi and remain Hi as they should, but I noticed that the IRQ lines of both chips (6502-4 & 6532-25) are both Lo , and remain that way until the test routine is completed by pressing the button twice. This is different than when the test routine is innitiated once the board is running properly - in this case the IRQ line remains Hi.

4) If I boot the Pin with the RESET line (6532-34, or 6502-40) jumpered to ground and kept Lo for a few seconds and then remove it - RESET goes Hi - The Pin boots perfectly, and the sound works perfectly. NO TONE OR TEST ROUTINE IS INNITIATED! My conclusion from this experiment is that the board is capable of booting up properly with a much longer "Lo delay". Maybe the fault lies with the actual delay circuit.

What I need is help understanding exactly how the delay circuit works, and how the individual components create the Lo, delay, and then latch the Hi signal to keep the processors running properly. Innitially, I thought that it was the charging of C3 that created the delay, but increasing it's value to 470uF made no difference in the booting success. After some reading of the data sheets, I'm wondering if it's actually the "switching time" of the 2 schmitt triggers U1A & U1B that's responsible for the delay. And I'm not sure of the purpose of CR1. Any help would be greatly appreciated.

5) Caps C3, 38, 39, 40, & 41, have all been replaced, and all board voltages are spot on. The pull up resister (DIP array 4116R-002-222) on the 6532 RIOT test pin also seems to be within spec at 2.2 K OHM.

Sorry for the long winded description, but this one really has me stumped, and any help would be greatly appreciated. I've attached a picture of the reset circuit for reference. It's deceptively simple.

Cheers, Amnesia

20210608_065402 (resized).jpg
#2 4 months ago

U1A-Cs are inverted Schmitt Triggers.

What the blank is a schmitt trigger?

In logic if you need just a snappy pulse on a logic 1 threshold and on its duration (unstable logic 1 but still a pulse) then you use a schmidt trigger.

It cleans your logic pulse now the dot on an input or output simply inverts that state.

What in this case the trigger A ouput goes low when the input high threshold is reached.

The trigger B goes output goes high when its input fron trigger A goes low.

I should do a logic write up.

#3 4 months ago

I was having an intermittent issue with a MA-216 in a spirit(speech/sound) but no speech chip. I fixed it by replacing the C3 cap. I figured my problem was no reset signal since it was more likely to work if it had been on for a few minutes.

If you replace this cap be careful the top trace likes to lift.

#4 4 months ago

Chedder hit the place you want to look.

R31 / C3 is your RC time constant and determines your reset period. The larger the capacitor value - the longer the recharge time.
Make sure the replacement C3 was 47uF and not 4.7uF. Also make sure it isn't an old NOS capacitor that could already be dried out resulting in a reduced cap value.

Also - if diode CR1 is shorted, you not get a valid result pulse -- will instantly end reset cycle.

2 weeks later
#5 3 months ago

Hey guys,

Thanks for all the replies!

My thoughts exactly. I replaced C3 with a brand new 47 uf cap, and even tried a brand new 470 uf cap but it made no difference what so ever. And yes those top traces are a piece of s###!!! Needed a little board repair after it lifted.

I wondered about CR1, but I was hesitant to desolder anything else. I guess that is my next step as its the only thing left to try. My understanding is the reset circuit stays low for a period of time (determined by time constant of C3 & R31) and then goes Hi -and stays Hi - to run. I don't really understand the function of CR1 in the circuit. Can anyone explain it to me?

Also, I ordered a new 6532 RIOT - I read somewhere that they can become "flakey" over time and can give some weird intermittant symptoms when they are failing. It should be here in a week. I also ordered a new processor (6502) even though its unlikely the problem.

BTW, for anyone recapping one of these boards, C38, 39, 40, & 41's leads are huge and have been crammed into tiny holes on the board. Even after using my HAKKO FR-301 (which I have become quite proficient with) it was almost impossible to pull the solder free leads out of the tiny holes without wrecking the traces!

Since game works perfectly after "power cycling", I'm certain there isn't a major component failure which makes it all the more frustrating.

Wow! What an adventure.

#6 3 months ago
Quoted from Bmad21:U1A-Cs are inverted Schmitt Triggers.
What the blank is a schmitt trigger?
In logic if you need just a snappy pulse on a logic 1 threshold and on its duration (unstable logic 1 but still a pulse) then you use a schmidt trigger.
It cleans your logic pulse now the dot on an input or output simply inverts that state.
What in this case the trigger A ouput goes low when the input high threshold is reached.
The trigger B goes output goes high when its input fron trigger A goes low.
I should do a logic write up.

Yup, I totally get all that. And U1A & U1B (inverted Schmitt Triggers) both test out working properly with my logic probe, so I don't think there is a problem with U1.

#7 3 months ago
Quoted from amnesia:

BTW, for anyone recapping one of these boards, C38, 39, 40, & 41's leads are huge and have been crammed into tiny holes on the board. Even after using my HAKKO FR-301 (which I have become quite proficient with) it was almost impossible to pull the solder free leads out of the tiny holes without wrecking the traces!

No kidding, BH's sound board was the first classic Gtb board I've ever recapped and I had to do some patching up of ripped-out traces. Afterwards it was recommended to me to heat the solder pads and bend the leads straight first before pulling the caps. And to add solder if needed during the process.

#8 3 months ago
Quoted from G-P-E:if diode CR1 is shorted, you not get a valid result pulse -- will instantly end reset cycle.

I can't for the life of me figure out what the purpose of CR1 in the reset circuit above is. I keep staring at it, but it doesn't make any sense to me (with my limited electronic knowledge). Does anyone know why its there, what it does, and how it does it?

Thanks, amnesia
20210608_065405 (resized).jpg

#9 3 months ago

CR1 prevents the IC input from exceeding the VCC threshold (+5V). The only time you will see this happen is during power down where the +5V line will go low faster than C3's voltage decay.
This will cause U1A input to follow the power down voltage rather than hold the U1A input high until C3 discharges through U1A and R31.

Hey there! Got a moment?

Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside