1981 Ms. Pac-Man restoration

[Finalheaven]

Hello, I've been real excited about posting on these forums for awhile. I recently acquired an old Ms. Pac-Man cabinet and, from what I can tell, it seems to be an original. The detail on the sides and coloring on the back are all printed directly on the wood, i.e. it's not a decal. It also has a serial number etched into the wood on the back that reads 31888. Also, when you look at the pictures that I'll link later on, you'll see on the inside and be able to tell right away that this thing really is close to thirty years old.

I'm posting here because I have a deep love for arcade machines in general, but especially for Ms. Pac-Man. I'm not a collector by any means, and this will probably square up to be my one and only machine. That being said, I'd like to get real intimate with the inner workings of the beast, which means that rather than go to a repair service I'd like to fix the thing up myself.

Some background on me: Years ago, I spent two years in engineering school, learning the ways of an engineer (I eventually left it and decided to go into finance). Currently, I build computers as a hobby for myself and friends and family. Likewise, in college, I would mod my playstation and playstation 2's, along with my friends', which required some light soldering. Anyway, I still have a decent understanding of electronics and computing, while my soldering is probably rusty.

What I'd like to do is find someone or a forums (much like these), where I could get some advice and instructions on how to get this thing working as I intend it to. I'll need some refreshers, but I'm a quick study and will catch on just fine. I don't expect any hand-holding, but an expert on these old-school cabinets would be excellent. Now, on to the machine.

Current state:

http://www.dragonhooker.com/images/pacman/front_display.JPG
So, I drove the machine 100 miles and wheeled it into my apartment. Just the looks and shouts I got on a highway from having a station wagon full up with Ms. Pac-Man goodness made it worth it. When I got it home and turned it on, I got what looked like a constant reset screen: a wall of white text, then colored text, then a grid, rinse and repeat. This screen was also too far up on the CRT and the image slanted upwards on the far right edge. First thing I did was make sure the little PCB's were seated on the main PCB.

http://www.dragonhooker.com/images/pacman/main_pcb.JPG
Also, the aux pcb had jumped out of it's seat, so I put it back where it belonged.

http://www.dragonhooker.com/images/pacman/aux_pcb.JPG
After that, the game would play fine and is on free play mode.
I fixed the video issues by using the knobs on the end of the CRT and on the CRT's PCB.

http://www.dragonhooker.com/images/pacman/monitor.JPG
While the game is playable, it does seem to be susceptible to intermittent resets. Sporadically, without warning, the game will just reset and return to the free play screen. This hasn't happened in awhile though, and it seems repressing the mini-PCB's into the seats stops the problem. The marquee light was also not functioning, so I tried replacing the bulb, when that didn't work, I also replaced the fluorescent starter, and now the marquee works. The joystick is old and didn't operate well, so I unlatched the control panel and used some duct tape to adjust the connections. It works better now, but I have a new grommet and panel overlay on the way regardless.

To be thorough, I also went ahead and purchased 2 5AMP fuses, 2 1AMP fuses, and 2 2AMP slo-blos. I tried replacing on the fuses, because I could see that some of the fuses had blown (I think one of the 7 Volt 5AMP fuses, if not both). When I had replaced all the fuses, the game actually wouldn't work, and returned a black screen with a random number or letter here and there. I replaced all the fuses back with the ones that the game arrived with and it worked again:

http://www.dragonhooker.com/images/pacman/mech_assy.JPG

List of TO-DO's:
1. I'd like to figure out and solve the random reset problem, even though it may be gone now.
2. Figure out why replacing blown fuses stops the game from functioning.
3. Reconnect the coindoor and freely switch between the free play mode and the attract mode, so I can watch the attract mode when there are zero credits in the game.

http://www.dragonhooker.com/images/pacman/coin_door.JPG
4. The game currently runs at an accelerated rate. I'm guessing this is because there's a speed rom installed? Also, if you look at the earlier posted pic of the front display, you'll notice a Zoom button. The Zoom button is not connected:

http://www.dragonhooker.com/images/pacman/zoom_button.JPG
I would like to reconnect this zoom button so I can control the speed, and at least be able to return the game to the original, normal speed.

MISC. TO-DO's:
1. Solder the difficulty jumper so I can play the harder version of Ms. Pac-Man, preferably with a toggle switch so I can switch between the two. I indicated the jumper that's mentioned in the Ms. Pac-Man operating manual (I have a 1982 original that came with the cabinet) in my picture of the main_pcb.
2. Rewire the volume knob located on the main PCB so I could relocate the dial to make it accessible from the outside.

Any and all advice would be appreciated. I realize I have a lot of work ahead of me, and will consider myself an apprentice to anyone who can guide me through any of these steps. I'm willing to correspond via this thread, PM, IM, email, phone and if you're in my neighborhood of Belmont, MA, in person.

And lastly, thanks to anyone who read through this entire thing. I truly appreciated it.

- Alexi

EDIT: I've resized the images and included them in the post. If you want to read my comments in the pictures, you'll still have to look at the full-sized versions. I also attached as many of the original-sized images as attachments for those who don't like to use outside links. Thanks for the tips. I'm working on getting more pictures of the full cabinet up ASAP.

[saleem]

why dont you just resize your photos and also include them to be shown on the posts you make,i hate having to follow links as it spoils the experience,after all,your experience is like ours also.we want to feel what you feel and get an high off your project/restore.after all this is where entusiasts get there daily buzz reading and looking at pics.

but,nice score,show us the pics in the post,just resize them so it fits on a 22"monitor,best is to resize for 15" and the reader can expand it.

want more pictures,sides,damage etc etc dont hold back so we can see what its like now and what it looks like as it gos along.
 

[Spyridon]

Post your pictures directly in this thread.  We are a lazy bunch and don't like jumping to other sites to look at the pics.

[RayB]

Dragon Hooker eh? 

[Finalheaven]

Dragon Hooker eh? 

You caught me.

[Finalheaven]

So, I've updated the original joystick with a new four cherry switch joystick system and used quick disconnects rather than solder. Unfortunately, Ms. Pac-man is still stuck in the downwards direction, i.e. when the game starts, her first move is down rather than the usual left. No matter what movements I make, she always returns to heading in the down direction. I looked for shorts along the down wires, but found nothing. Any ideas?

I figured out how to get the machine into normal, attract mode. I just had to set the dip switches #1 and #2 to coin play. After doing this, I tested the coin door. Amazingly enough, both coin slots fully function, accept quarters, count each quarter that passes on the counter, awards credits, and returns stuck change. I'm still waiting on the coin door light bulbs, so no word on that yet. I also found out that the red button and black switch behind the counter box awards credits. Basically, if I press the red button, it awards one credit but does not trigger the counter. The black switch is still a mystery to me. I also can't seem to figure out how to get the system into self-test mode.

[Level42]

So, I've downgraded the original joystick with a new four cherry switch joystick system

Corrected.

[RayB]

Trying holding the black switch while powering on. Hold it for several seconds. Might put it in test mode.

Now I have to second Level42's post... why did you replace an original joystick with a microswitch one?! Did you even at least bother to make sure it's a 4-way stick?

I hope you didn't throw out the original.

Re the moving down, obviously something related to the down input is making a connection somewhere. Disconnect both wires from the down button and see what happens then. If it still makes her go down (    ) then the short is somewhere else, maybe at the board connector. I'd then disconnect the entire control panel (assuming you still have a molex connector making that possible) and see if she still moves down.

[Finalheaven]

I did keep the original and immediately regretted taking it out after the fact. I'll put it back in. I'll try those tips next.

[Level42]

Mistakes is how we learn Trust me, I've been there and still make plenty mistakes !

The Pac Man Joystick is IMHO one of the best arcade controls ever. It's also sturdy and relatively easy to get into a great like-new shape.
Definitely worth it.

[Finalheaven]

Mistakes is how we learn Trust me, I've been there and still make plenty mistakes !

The Pac Man Joystick is IMHO one of the best arcade controls ever. It's also sturdy and relatively easy to get into a great like-new shape.
Definitely worth it.

Yeah, agreed. I was worried more about the stuck down situation. It seems my error in judgment is absolute though. I had the spare parts in one plastic bag, and trash in another plastic bag. I ended up throwing both out. Yup, that's right. I mistakenly threw away an original Ms. Pac-man joystick.  <---- x1,000,000^42

I'll have to look for another original somewhere.

In the meantime, in between long bouts of kicking myself in the butt, I've been checking the wiring. The Brown/white striped wire is the lead for the down arrow. I unplugged that wire and she still went down. I disconnected the control panel molex and she still went down. I followed the brown/white wire all the way back to the long black pin connector on the PCB. I disconnected the pin connector and looked for anything that might indicate a short anywhere near that brown wire and found nothing. I also disconnected the molex leading to the coin door, just in case, and she still went down. I'm guessing that the short must be occurring on the PCB itself? I followed the pin the brown wire connects to on the PCB, to see if any of the nodes on the PCB relating to the down direction were located near a sloppy solder of some sort. I found nothing. So, I'm stumped. I doubled checked the wiring schematic in case I was missing something, but it's pretty straight-forward. My board does have the 12/7/81 revision by V. Caporusso that adds the credit multiplier PC and the fluorescent light.

Back to the drawing board.

[Finalheaven]

So, I used the multimeter to measure the pins on the 74LS chip responsible for joystick movements.

I found the pin layout (if that's the right term) for the 74LS367 chip responsible for up-down-left-right. I also found this great site (http://lawnmowerman.rotheblog.com/logicprobe.html) that basically lays out how to measure it.

I used my schematic and found that pins 14 to 13 are responsible for Down on the 74LS chip at location 8E. I also learned pins 10 and 9 are responsible for Up. Basically, I measured the voltage between pins 14 and 13 and found they were always at 4 volts (I had my multimeter set to AC10V). Pins 10 and 9, 6 and 7, and 4 and 5 were 0. I then measured pins 10 and 9 while the up switch was depressed and it read 4 volts. When turned off, it went back to 0. So pins 14 to 13 are always on. What does that mean? Bad 74LS chip? I also learned R45 and C20 are the resistor and capacitor responsible for down, respectively (220 ohm, .1 microfarad). I have no clue if the Res. and Cap. are contributing to the problem.

I think there is light at the end of this tunnel.

[Finalheaven]

Picture of the cabinet and screen (while it's working):

http://www.dragonhooker.com/images/pacman/cabinet_back.JPG

http://www.dragonhooker.com/images/pacman/cabinet_front.JPG

http://www.dragonhooker.com/images/pacman/cabinet_screen.JPG

[saleem]

are you going to keep the art or are you going to give it a stencil overhaul?

it certainly looks like a sturdy cab.
 

[Finalheaven]

are you going to keep the art or are you going to give it a stencil overhaul?

it certainly looks like a sturdy cab.
 

Despite the sun bleaching, I think I'll keep the original artwork. It may be faded, but that's part of its authenticity.

I have two steps now. I'm going to replace the 74LS367 chip (install a socket and then the chip in slot 8E) and I'm going to rewire the pin connector with one of these: http://www.twobits.com/Pacmend/.

Anyone have any suggestions for which IC 16 pin socket to use for 8E? There seems to be a lot of choices. Overwhelmingly so.

[Finalheaven]

So here's the update:

I attached an edge repair kit to repair the burnt edge connectors. I removed the 8E IC and installed a 16pin socket and a new 74LS367 chip.

I'm guessing I did all the work right, because the game still boots up, but I'm still plagued by the same issues. Ms. Pac-man still constantly moves down, there's a scrolling dark vertical band that scrolls across the screen from right to left, and if I try to play a game after a few seconds the game corrupts into several changing shapes on a black background and then resets back to the starting screen.

I'm going to try removing all the removable IC's and cleaning their pins while checking the traces on the 8E socket. After that, I'm stumped.

[SirPeale]

What feeds that chip? 

http://www.datasheetcatalog.org/datasheets/105/488554_DS.pdf

[Finalheaven]

Sorry, I don't know enough to answer that question. 

[SirPeale]

Nor I, TBH.  But let's look at this logically:

1) the 74LS367 is an input chip.  Those inputs are on all the time.   Replacing the chip has not solved this issue. 

2) according to the schematic I linked above, other than incoming voltages there are only two pins (IIRC, I'm not looking it it right now) that don't have anything to do with inputs.  Trace those leads back to those chips and see where they go.

[Finalheaven]

The only input that seems to be constantly on is the down direction. The rest work perfectly. My bigger problem now is the instability due to the intermittent resets. If I can get the game to fully load and make it to the credit screen, I can start a game and play for a few seconds before it resets.

I'm about to replace the fuse block and the 6H and 6J eproms, as they're pins are really brittle, etc. I may replace their sockets too.

[SirPeale]

I read here:

http://www.mikesarcade.com/arcade/repairs/mspacman.html



That one of the resistor packs was broken in pieces when he unsoldered it.  Check it.

[Finalheaven]

Here's the current update:

I've fixed the random reset problem. I'm not sure if it was related to bad eproms at 6H and 6J, but after noticing their pins were brittle and falling apart, I replaced both eproms. This was after fixing the edge connector which didn't seem to change my status (random resets, vertical hum bar, down direction). Regardless, after updating 6H and 6J and soldering a new edge connector, the resets were gone and I was still stuck with a vertical hum bar and Ms. Pacman constantly going down (she'll accept directional inputs, but then still head down afterwards):


After that, I wanted to clear up the vertical hum bar problem. I read the paragraph about it on lawnmowerman's pac-man page, and I knew I had already fixed the edge connector, so I replaced the fuse block. Vertical hum bar is now gone:


I'm SO close to being finished, but I still have this constant down direction problem. She's still controllable, but always heads down by default and will return to that direction if no other input is given. Since I've already replaced the 74LS367 chip and socket at 8E, my last thought is a short on the board. I did notice when installing the new 16 pin socket that the traces around the sockets holes were very close together and unprotected by the usual green PCB stuff (I forget what it is, but it's not conductive). I have a logic probe coming tomorrow as well. I'm hoping a reinspection of the underside of the PCB around 8E and a logic probe will help me uncover my final problem. Then I just need to get the coin door lights working (that should be a cinch after this stuff).

[saleem]

i dont know about this one.but,have you checked or replaced the down microswitch on the control stick,pretty obvious isnt it,worth a shot.
 

[SirPeale]

Did you check the resistor packs?

Also check the capacitor on that line.

I started a thread @ KLOV: http://forums.arcade-museum.com/showthread.php?t=137534

[Finalheaven]

Thanks a lot for that. I should register for those forums as well.

I have done a thorough test with the down direction (which is the top switch) disconnected and the entire panel's molex disconnected as well.

I believe the resistor on the down direction path is R45 (from memory). None of the resistors look fishy though, but looks can be deceiving. As for the capacitor, checking the manual, it looks look... (checking) C20 is the capacitor responsible.

In the other forums, is that the capacitor he is suggesting I disconnect?

While I'm not through the storm yet, it makes it all bearable with help from you guys, so thanks for that.

[Finalheaven]

That did it! She moves perfectly now. Interestingly enough, the game is playing perfectly while the C20 capacitor is still disconnected.

Regardless, I would like to replace the capacitor. Which capacitor should I buy exactly from radio shack?

Also, I'm noticing *slightly* too much read on the first two stages which gives the maze walls a sort of pseudo-3D effect by bleeding on the top and right side of each wall segment. I assume this is not intended and I should reduce the color via the monitor controls.

[SirPeale]

They may or may not have them @ RS.  YMMV.

I was talking about resistor packs.  They look like a one-sided centipede.   But as you seem to have found your issue I guess they're fine!

[Finalheaven]

It's running pretty well now. I notice some movements get missed and I'm wondering if this is due to the disconnected capacitor. What type of capacitor should I get? Like what rating?

[SirPeale]

It's some kind of ceramic cap, and I'm not sure of the rating.  Someone in the other thread mentioned .1 uf, but if that's actually the value I can't say.

[Finalheaven]

All done. Everything's in working order. Coin door lights completed. That was a lot worse than I expected it to be. Anyway. My last project will be to perhaps wire a difficulty switch for the difficulty jumper located on the board and add a sound switch somewhere.

Thanks for all the help. That was awesome/nuts.

[Finalheaven]

Here's a weird question (and possibly a dumb one):

I have the normal speed and fast speed chip for 6F. I have the fast speed chip installed currently. Is there a way to perhaps create some sort of setup that will allow me to toggle between the two chips? I heard of the 4-in-1 chip, but that also includes cheats.

Any thoughts?

[SirPeale]

If you want to mess around with the 4-in-1 before doing any installing, it's supported in the latest version of MiSFiT Mame.

[Finalheaven]

Is anyone up to speed on hardware speed hacks for the ms. pacman machine, like described here? http://lawnmowerman.rotheblog.com/#speedUpsHard

When I got my cabinet, I noticed there was a Zoom button installed on the control panel. The wires were cut, but it looked like at one point it had made the game speed up.

Or, perhaps safer for the system, how about piggybacking the 6F chip and installing a switch to toggle between enabling either output pin? Any advice there? I may go with this option since I already have the chips for 6F and some spare sockets. All I would need is the switch and a wiring scheme.

Does anyone know the OE pin for the 6F chip? Thanks.

[Finalheaven]

So I created this thing. Using jumpers, I basically connected all but two pins of each 24pin Eprom in parallel. There are three sockets on the other side: one socket for the ribbon cable, one socket for the fast chip, and one socket for the normal speed chip. Pins 18 and 20, the OE pins according to the Ms. Pacman schematics, are connected to a DPDT switch (you can see the backside of the switch at the top right of the picture). The idea is to switch between the fast chip and the original speed chip via the switch. I know of the hack Eprom but I didn't want invincibility on my chip or a simple speed up for Ms. Pacman.

I connected the switch and when I turn on Ms. Pacman, I get an image of random colored numbers and letters with an orange here and a pretzel there. I still have some kinks to work out. I've checked that none of the wires have snapped (already three have and I had to resolder them. 30 gauge wire is a effing pain). I used a multi-meter to continuity test the pins and sockets and switch. Everything checked out. I'm going to test for shorts now and test the pins on the end of the flat ribbon cable with the pins on the eproms, right at the shoulders.


http://www.dragonhooker.com/images/pacman/eprom_switch.JPG

[RayB]

I think the way I have seen it done is to piggy-back the chips, and bend a couple specific legs up (maybe the ones that take power?) and you link those to a switch (2 or 3 position switch depending on how many chips you wish to support).

This is all fuzzy memory of how I *think* my Galaga's fast shoot chip is set up. Think about it, no soldering of dozens of little wires!!

[Finalheaven]

I think the way I have seen it done is to piggy-back the chips, and bend a couple specific legs up (maybe the ones that take power?) and you link those to a switch (2 or 3 position switch depending on how many chips you wish to support).

This is all fuzzy memory of how I *think* my Galaga's fast shoot chip is set up. Think about it, no soldering of dozens of little wires!!

I wanted to create the switch board so that I can swap eproms and not worry about bending pins. When I was pulling the 30 year old eproms out of the PCB some of the legs fell right off, and that was with perfect extraction technique. I'd rather screw with little wires and get that right and leave the eproms untouched. Sockets and wire is cheaper for me than finding 2532's, plus I really enjoyed it.

As for the voltages, I read a few data sheets regarding 2532's and discovered pin 21 is Vpp, pin 24 is Vcc, and pin 12 is Vss. From a TI datasheet (http://www.downloads.reactivemicro.com/Public/Electronics/ROM/2532%20EPROM%20Data%20Sheet.pdf) I read that Vpp is +25V power supply and the Vcc is the +5V supply and Vss is the ground.

In the old Ms. Pac-Man schematic, There's no record of the 24 pin on 6F, but it does show pin 21 as Vpp with a straight line pointing out and then ending at a diagonal line. Any thoughts on which pin I need to open in order to disable the eprom?

[RoyalScam]

First off, let me say excellent progress on the restore!  I have absolutely no experience repairing Ms. Pac-Man or any other arcade machine for that matter, but Vpp is usually the programming voltage supply for an EPROM. Should that voltage be present, I would think that it would be very possible to randomly reprogram, or even irreparably damage your ROM.   I would leave that pin unconnected.  Should you want to verify this with the original mainboard, use a dc voltmeter set to read in excess of 25 volts and check to see if Vpp is actually being supplied to the EPROM.  I would guess not.

Regards,

Scam


<edit> Thinking about it a bit more...That pin may be held high Logic1 or +5volts or even held low Logic0 or GND, easily verified with the voltmeter.

[Finalheaven]

First off, let me say excellent progress on the restore!  I have absolutely no experience repairing Ms. Pac-Man or any other arcade machine for that matter, but Vpp is usually the programming voltage supply for an EPROM. Should that voltage be present, I would think that it would be very possible to randomly reprogram, or even irreparably damage your ROM.   I would leave that pin unconnected.  Should you want to verify this with the original mainboard, use a dc voltmeter set to read in excess of 25 volts and check to see if Vpp is actually being supplied to the EPROM.  I would guess not.

Regards,

Scam


<edit> Thinking about it a bit more...That pin may be held high Logic1 or +5volts or even held low Logic0 or GND, easily verified with the voltmeter.

I did that and found Vpp was not being supplied 25 volts. Pins 21 and 24 were being supplied +5 volts. I rewired the switch so that pins 21 and 24 would be operated by the DPDT switch, but it still didn't work. I was getting artifacts. At this point I've settling for the original speed rom and closing her up.