Looking for a 0066-117xx chip

[Scottacus]

I am restoring a Gorf game and one of the audio channels is toast.  The audio 2 pin is melted in the molex connector and when this fried connection is bypassed only garbage comes out of the audio 2 channel.  I did a wire wrap onto the audio pin of that 0066-117xx chip and ran it through an amp to bypass the rest of the audio circuit and that gave me the same garbage so it is the chip or something upstream that is bad.  The other audio channel has the same model of chip and it receives the same audio input and works fine so I think the problem is not upstream but in the chip.

This chip seems to be custom made for Bally/Midway and is a general purpose I/O chip with 40 pins.  It was used in Astrocade as well as Gorf and perhaps somewhere else.  If anyone has one in their pile of parts I would be very interested in it.

[SavannahLion]

I have no such IC available but if it helps in your search...

IIRC that chip was also used in
    Extra Bases
    Robby Roto
    Seawolf II
    Space Zap
    Wizard of Wor

I believe it was also used in one or more pinballs, but I can't recall what they are off the top of my head.

There was mention of someone supposedly reverse engineering the Astrocade in FPGA for the Spartan 3E, but I can't seem to find the source code or even the binary on the site

The Bally I/O chip doesn't have as much love and attention as the Atari TIA so I haven't seen a schematic or even a clear cut explanation of what the IC actually does. Well... I know what it does, I mean a painfully detailed breakdown like what the infamous TIA received. It would be wise if someone did since I've read the failure rate on those Bally chips are pretty high. 

Hopefully that extra information will help you broaden your search. Might luck out and find a cheap spare board on eBay with that IC. Good luck...

[Scottacus]

Thanks, that helps!  Since there are two of these ICs on the board I might have some additional options. 

I tried disconnecting the audio 2 wire, figuring that the audio 1 would drive both speakers but it wouldn't put out any sound, well I shouldn't say any sound, it made a faint sound if you stuck your head into the game. 

If I leave the audio 2 (bad) and audio 1 both hooked up I get a bad mix of the two out of each speaker with plenty of volume in my current mono mode setup. 

The audio 1 (good) signal has a little high pitched overtone with the speech generation that a casual listener probably won't hear but it bugs me so I will wire wrap off of the audio output of the good (audio 1) 0066-117xx chip to see if the overtones are from the chip or the amplifier circuits downstream.  Depending on the outcome of that I might just disconnect (bend out) the bad audio 2 0066-117xx audio out pin and wire wrap the good audio 1 pin to send its signal down both the audio 1 and 2's amplification paths.  Isn't trouble shooting fun!

[Scottacus]

error

[Scottacus]

I tired the wire wrap from the good chip's audio out and ran that output through my guitar amp and got wonderful sound but only half of it.  I was missing the sound effects for Galaxians falling, space invaders stomping, laser ships shooting...

I now see that the two 0066-117xx chips on the board are not marked exactly the same and are probably slightly different.  Here is the pinout for the chip

         
             +-----+--+-----+
      GND |1    +--+    40| MXD7
       SI0 |2                39| MXD6
       SI1 |3                38| MXD5
       SI2 |4                37| MXD4
       SI3 |5                36| MXD3
       SI4 |6                35| MXD2
       SI5 |7                34| MXD1
       SI6 |8                33| MXD0
       SI7 |9                32| SO0
    POT0 |10              31| SO1
    POT1 |11              30| /INT
    POT2 |12              29| SO3
    POT3 |13              28| SO4
DISCHG |14              27| SO5
 MONOS |15              26| SO6
         Φ |16               25| SO7
 /RESET |17              24| AUDIO
    TEST |18               23| /RFSH
  /IORQ |19               22| Vgg
       /Φ |20               21| Vdd
             +--------------+

Vdd: +5v
Vgg: +10v

Both audio chips use the MDX 0-7 pins and although I'm not sure what MDX stands for, I think this is where the 8 bit data comes in to drive the sound effects.  Both chips get fed the same data (all these pins are tied together in series MDX0 on X2 connects to MDX0 on X3) but they must each interpret it differently to put out half of the sounds (ie one makes the quark laser sound and the other makes the gallaxian sound).  That is my current working theory.

The speech generator chip feeds in after the pin where I was taking the sound out so the high pitched overtones are coming from somewhere downstream from the 0066-117xx...

Any advice on this would be greatly appreciated!

[SavannahLion]

If the chips are identical but are behaving differently, then the only interpretation difference is what's going in. You'll need to look at the schematic or study the circuit a bit more to get a better idea of what's going on. I took a peek at the Gorf manual to get a better idea of the I/O and noticed that it lists an alternate part number, 2720. A search yields virtually nothing on this particular IC but does turn up this thread from 2009. http://forums.arcade-museum.com/archive/index.php/t-104475.html
And this thread from 2007. http://forums.arcade-museum.com/showthread.php?t=83034

Interesting to note that the 2720 is listed as an alternate part number as well as MA 0066 070BX

[Scottacus]

Thanks SavannahLion, I appreciate your taking time to look into this!

I am not an electronics expert but it looks to me from the schematic and looking at the tracings that the same signal is fed into both 0066-117xx chips but they do have slightly different part numbers.  One chip puts out great sound and the other garbage.  When I watch videos of folks playing this game I see that I am missing several sounds.  A couple of questions:

If the same signal is being fed to both chips (assuming they are identical) how is it that they would interpret the data differently and make different types of sounds? 

Is something else controlling how this 8bit data is interpreted and output?



X2 is my left channel bad chip and X3 is the right channel good chip, the audio out to the amplifier circuit is pin 24 on the bottom of each chip and the MXD inputs are pins 33-40.  The bad chips SI 0-7 are tied to ground while the good chips are sent to U 1-4

[SavannahLion]

Thanks SavannahLion, I appreciate your taking time to look into this!

I am not an electronics expert but it looks to me from the schematic and looking at the tracings that the same signal is fed into both 0066-117xx chips but they do have slightly different part numbers.  One chip puts out great sound and the other garbage.  When I watch videos of folks playing this game I see that I am missing several sounds.  A couple of questions:

If the same signal is being fed to both chips (assuming they are identical) how is it that they would interpret the data differently and make different types of sounds? 

Is something else controlling how this 8bit data is interpreted and output?



X2 is my left channel bad chip and X3 is the right channel good chip, the audio out to the amplifier circuit is pin 24 on the bottom of each chip and the MXD inputs are pins 33-40.  The bad chips SI 0-7 are tied to ground while the good chips are sent to U 1-4

edit:

There's no way I can read the line names and half the pin numbers on there but assuming that X2 and X3 are the same chips, it's pretty clear they're not in the circuit the same way. The XOR gate and buffer stands out on X2 whereas there's an entirely different circuit in X3 Some of the pins also look like they're left floating (X2 looks like it has fewer pins than X3). Anyways, The XOR gate goes into what looks like the interrupt? Both IRQ's appear to have different notations or that might a smudge on the diagram. 

I can't tell what U1 throught U4 are. Shift registers? no.... the outputs look like they're labeled X1-X4 and Y1-Y4.... Dual decoders? MC14539B maybe? That would work.... Probably hints at what's going on with those two. One IC pulls more of it's share of work than the other.

It is entirely possible to feed the same data into two identical IC's and have different ouputs based other inputs into that IC. On the surface, it looks like the behaviors shouldn't change but when you start adding things like XOR gates, NOT gates and a whole other slew of circuitry, that's when you start having fun

[Scottacus]

Thanks again for your taking the time to look at this and giving me your input! 

The scan of the schematic is not the highest fidelity but I am very glad to have it.  I have a different set of boards coming soon so I'll see what happens when I do a "brain transplant" to the game!

[SavannahLion]

Apparently, such "high fidelity" was pretty common back in the day. 

There's a forgotten art to drawing and reading them. Is that part of the GORF schematic?

[Scottacus]

Yes it is, the original PDF can be found at http://www.arcade-museum.com/game_detail.php?game_id=7983

[Scottacus]

I jus got a new complete set of boards and of the six boards all work except the pattern board and one RAM board.  The game board now has two good 0066-117xx's on it and it plays the game in glorious stereo now!

I did a little experimenting and found that if you put the bad 0066-117xx in the X3 slot and the good one in the X2 slot the game will not boot up.  I was very fortunate to have them where they were when I started out because I doubt that I would have been able to trouble shoot a completely dead game and figure out that the 0066-117xx in X3 was the problem.

So in summary, if you get a dead Gorf game and want to trouble shoot it, one thing to try is switching X2 and X3 on the odd chance that a bad 0066-117xx in X3 is the cause of your problems.

[SavannahLion]

Nice! 

I find it interesting that I read several accounts of X2 burning out, but nothing on X3.

Looking at the schematic, I have to wonder if there might be a bad design choice accelerating the death of X2.

I don't know enough details about the 0066-117xx to make any definitive statement and I can't find jack ---steaming pile of meadow muffin--- about the operation of that IC specifically (I'm spoiled, there's entire damn books written about using the TIA but only the most vague information about the 0066-117xx ). A lot of digging turned up http://www.ballyalley.com/ that looks really promising.

I wish I had more time so I could really dig into this. It would be very interesting to learn more about that particular chip.

[Scottacus]

Here is some more info about the 0066-117xx chip that I found on a Midway publication about the card rack system.

[SavannahLion]

Interesting... Must mean the MUX lines are capable of going to a Z state to share the bus.

I've never seen the phi symbols on the pinout of an IC before. At first I did a bit of research but couldn't figure it out. None of the older docs I have make use of Phi either.  I eventually found an old spec sheet for the CD4060 that looks like it uses Phi for a clock out and an inverse and phi for a clock in.

I'm not sure where the clock is fed into on this particular IC but I would bank that's what it means.

[Scottacus]

This helps explain why X2 can be bad and the game will still run (thank heaven for that or the Gorf would have been DOA).  Since the SI pins are all tied to ground that chip must just be passing the sound signals through to the amps below.  The X3 chip is getting fed data through MUX (MC14539B's) attached to its SI pins so if that chip is down, some form of data is missing.



Looks like Phi and Not Phi are the CPU clock high and low.  I got that from a Space Zap manual so it must refer to the 0066-117xx chip since that game also uses it.

[SavannahLion]

Huh... I haven't looked at the Space Zap manual yet.

You have to wonder though, why is X2 getting burned out? From what I understand, it's not even running all of the audio but half. It's overheating for some reason. Is it physically higher or above the other components? Maybe it's getting cooked by the rest of the circuit below.

[Scottacus]

Good question!

Something VERY BAD happened in that circuit.  The audio channel pin on the output side of the amps that was connected to the bad 0066-117xx was fried, as in the molex was melted and the pin charred

Something put a lot of heat on that output wire...