Here you will find Repair tips, tricks and
other information I found very useful and you may, too. A lot of the stuff
is just thrown together at the moment. I'll sort it out and categorize it
later.
Williams/Bally Tempered Glass Measurments:
Playfield:
23 3/4" x 43" x 3/16" tempered glass NO LOGO **WIDEBODY**
21" x 43" x 3/16" tempered glass NO LOGO $28.15
Backglass WMS:
18 15/16" X 27 1/16" X 1/8" tempered glass NO LOGO
Pop bumper bulbs keep burning out? Have a bulb in a very hard to reach place? Try Buying some #444 bulbs.
These are heavy duty #555 bulbs with thicker filament and support to make the bulb resistant to vibration. Same voltage and brightness as the original.
When I was working on my alphanumeric pin, it was next to impossible for me to find this information. So I'm archiving it here for future reference and to share with my pinhead friends.
Williams Displays:
Gottlieb Displays:
Guns and Roses Magnet Board Schematics:
Does your WPC Williams/Bally game reset? Here's a collection of comments on the subject gathered from RGP Newsgroup.
Look at connectors J101 and J102. If either
of them are cooked it will
cause a reset. Recently someone posted that a cooked DMD driver board
can also cause a reset.
I had a very similar problem when I first bought my TZ. It turned out
that 2 of my boards had mounting screws that were loose/missing. Simple fix,
and it's never acted up again.
What is the line voltage at your wall plug? Most reset problems are due
to low line voltage- not many are due to bad bridges. These games are
designed to run at 117-120vac. When you get down to around 114 and
113vac, then reset problems start to occur.
Game won't turn on? Pulses on turn on? Hit both flippers and game
shuts down? Chances are you have a bad diode on the bridge rectifier
for the +5VDC supply. Here is an easy test where you can avoid
soldering UNTIL you are sure you know the problem.
You will need a 6A 100PIV diode, a voltmeter, and a few jumper wires
with clips.
Now find bridge rectifier BR2 on the Power-Driver board (top one under
heatsink). There are two wires near the top on the underside of the
bridge, the one to the left of center is the + one and you will be
clipping one of your jumper leads here.
Now hook this to the positive lead of your voltmeter and connect the
negative lead of the voltmeter to the metal braid wire running along
the bottom of the head (ground).
Turn on the game and see what DC voltage you get. It should be at
least 9.00VDC, if below that get your diode and another jumper wire.
Hook the jumper wire to the banded end of the diode and connect the
other end of the jumper to the junction of the positive lead for the
voltmeter and the jumper to the bridge.
Now take the diode and with the power on insert the other lead of the
diode into the edge connector labelled J101 in either pin 1 or 2 (top
two wires). Watch the voltmeter while trying the lead of the diode in
each connector. If the voltage rises when the lead is inserted in the
connector then the bridge rectifier has a bad internal positive diode
and must be replaced.
IF no change then remove the diode jumper to the junction of the meter
positive and jumper form positive of the bridge and then move the
jumper wire from the banded end of the diode to teh non-banded end and
connect the other end of the jumper to the ground braid wire on bottom
of the head.
Now use the banded end of the diodes' lead and insert it into J101
pins 1 and watch the meter, if no change then insert into pin 2.
IF still no change then you probably have a bad filter capacitor and
you can set the voltmeter on AC reading and see what you get. It
should be under 0.30VAC. If it is higher then try connecting a
replacement filter capacitor with the positive to the junction of the
jumper form the bridge and the meter, and negative to the metal ground
braid. If the AC voltage then drops the filter cap is bad. This is
very difficult to remove without damaging the circuit board. DO NOT
PULL IT OUT WHEN UNSOLDERING, it might be better to have a
professional help remove it. If you damage the board check Marvin's
site on jumpering the filter cap. Good pictures of the job.
I do recommend the use of 10,000 @ 16 or 25VDC capacitors to reduce
inrush current surge to protect the bridge rectifier. Williams went to
the 10K capacitors most likely for this reason. The in rush surge is
most likely the cause of bridge rectifier failure.
How to use a logic probe. This is mirrored from Mowerman. I found this extremely useful when I started board repair and if it should disappear from the author's page, we'll still have access to it.
{Insert picture of probe here}
A Logic Probe is one of the basic troubleshooting tools for common electronic logic circuits (TTL circuits, chips with the 74XX pre-fixes). Almost all video game circuit boards are built using TTL chips. These chips will occasionally fail and the Logic Probe is a tool that affords us a look into what may be failing. Logic probes are fairly inexpensive about $30.00 from Jameco, Radio Shack (Hurry up RS is closing these out now 5/03!) or the like. Logic probes have their shortcomings too...
Logic probes derive their operating power from the PCB under test, probes have 2 power clips Black for Ground & Red for +5VDC. Care should be taken in attaching the clips to safe voltages; you may want to first use a Volt Meter (Digital Multi Meter) to verify the hook up points have good voltages. (Actually if you are at the point of using a logic probe you SHOULD have already verified the board under repair has good voltages.)
Logic Probes have 3 LED's and some have a Piezo buzzer. The 3 LED's indicate...
RED High logic state a binary 1. TTL chips consider +2.4 to +5
VDC to be high.
GREEN Low logic state a binary 0 TTL chips consider 0 to + 0.8 VDC to
be low.
YELLOW Pulsing state (yellow &/or piezo buzzer)
Your actual LED colors may vary depending on make!
Notice the voltage range between +0.8 and +2.4 is undefined. Any steady
voltage state in this undefined range should not light any LED's on the logic
probe & is indicative of a problem such as overloaded/ weak output or missing
pull-up resistor.
Transistor Transistor Logic is the commonly used format of logic chips on
PCB from the mid 70's to present day. These chips may be identified by the
numbers printed on their top.
74XX Standard TTL family
54XX Military Grade TTL, made to higher standards against radiation, moisture
& failure.
93XX Early TTL family.
The TTL family has been improved over the years in speed & performance. These newer sub classes are identified by modifiers such as....
74LSXX Low Power Schoktty Type
74FXX Fast
74SXX Schoktty Type
74HCXX High Speed CMOS Type
All chips perform the same logic operations, a 7400 Quad 2 input NAND gate will perform the same as a 74LS00 but there is minute improvements in Propagation (speed at which the signal travels through the gate) and actual power consumption by the individual chip.
When replacing a TTL chip you should strive to match the original chip.
It is acceptable to replace a 74XX with a newer class 74LSXX but you should
avoid retreating back down the ladder of evolution and replace a 74FXX chip with
a 74XX for example (in this case you are replacing a high speed chip with a slow
speed chip & inviting problems).
40XX or
45XX
Complimentary Metal Oxide Semiconductors. This logic family is a step beyond the TTL family. It's most notable feature is that it can operate at much higher voltages (over +12 VDC) than a TTL and that it is ANALOG in it's performance compared to the DIGITAL nature of TTL. TTL's are considered digital as they have only 2 valid states Low (0) and High (1). CMOS have states that are proportional in nature. For example if you had a 4069 Hex Inverter, with a functional (not necessarily pin out compatible) equivalent of 7404; we know that in a 7404 a given gate input state of High (1) then it's output would be inverted or Low (0). The 4069 would be different, first it's input state could be a voltage range from 0 to the operating voltage, say in this example +12 VDC. If the input voltage for a given gate was +4.0 VDC then the output would be +8.0 VDC or the inverse proportion of the input relative to the operating voltage.
CMOS is useful for handling audio outputs (like on Pac-Man using the 4066 at 1N) or creative people have even used them to do quickie video inversion on Nintendo games for the color section.
It is possible to use a Logic Probe to troubleshoot CMOS chips if the
probe is designed to handle CMOS. Most probes have a switch to toggle between
CMOS and TTL chips.
Well now is when schematics become important. Schematics are the road map
for the logic signals that are traveling around the PCB.
Most schematics will have labels marked on them such as RESET
(Imagine that the line is drawn above the word RESET and not through it, HTML
does not support a bar over characters yet). The line over the word RESET
requires us to read it as "RESET NOT". This method also indicates what state it
is, normally the PCB trace that contains the signal RESET is
High, (1) or +2.4 VDC or greater. When the event is active, the Pac-Man board
decides it is time to perform a reset then this PCB trace will go Low and
initiate a reset function. The RESET is called a active low
signal meaning that the reset function is activated when the line goes low.
Schematic labels do not follow strict naming conventions. Chips sometimes will be marked as for inputs and outputs. The letter Q always indicates an output.
Schematics may also have arrows on given traces, this is to indicate the flow of the signal, if it is a input or output to your given chip. Generally schematics are organized so that the flow of a signal follows a logical convention. Inputs generally enter the circuit from the left or top and the outputs exit from the bottom or right sides. Signal processing generally is shown to proceed from left to right and top to bottom on schematics.
Now would be a good time to learn a bit more of the ins and outs of these chips. The TTL Cookbook by Don Lancaster will provide you with some insight into what actually is going on inside of these chips. ECG & NTE, both replacement chip suppliers, offer catalogs that include descriptions and pin out diagrams. Any of these sources or even manufactures' data books will give you some idea of which pins are inputs and outputs.
Online several chip manufactures have their data books, my personal
favorite at the moment is Fairchild (www.fairchildsemi.com). Texas Instruments
just recently sent out "freebie" pocket reference books with a listing of most
common TTL, pinouts, truth tables etc. Worth getting but it has an error on
74LS157!
Okay, so grab your probe & schematics. Since you have already checked the
Pac-Man board for voltage clip your logic probe leads to a power source, I like
using C2 or C3 both are moderate size electrolytic capacitors (470 mfd at 16VDC)
and are marked as to which end is positive/ negative. C2 is central to the
PCB. (If you are working on something else you may want to just use a rail
filter cap, most Printed Circuit Boards have the TTL chips arranged in rows and
columns, columns go up & down and are flanked by power rails with Ground being
the opposite end of pin 1 and Voltage (+5 VDC or VCC) being at the same end as
pin 1. Each of these rails usually will have a small "spike filter" capacitor
across them where you can attach your logic probe to power.)
Now test your probe by touching the tip to the positive side of C2. You should have a solid, steady {RED} LED indicating high (some probes will also have a tone). Your board is powered up right?
Next probe the negative side of C2, solid steady {GREEN} LED indicating low state.
Let us see if that RESET line is working. Probe the
RESET line at the Z80 @ 6B pin 26 (or the test connector pin 35
or the 74LS02 @ 7L pin 13 or the 74LS259 @ 8K pin 15 or the PCB edge connector
pin 6).
Your probe should indicate a High state, {RED} LED.
Hold your probe in place, watch the LED's and press the red button on
the Pac-Man board (SW1). You should notice the {GREEN} LED light as
RESET is activated by the low state you just witnessed & the entire
game will reset and cycle through the startup test routine as if you had just
turned the game on.
Notice that there is also a RESET signal shown on the schematics between
the 74LS161 @ 9C pin 15 and the 74LS02 @ 7L pin 12. RESET is the opposite state
of RESET and would act opposite of the states you just observed
previously.
WHAT not working?? Hmmmm, if your game is infact running but the
red SW1 has no effect, test your SW1 first ~ probe the 74LS161 @ 9C pin 9. This
should be high & go low as you depress SW1, if it doesn't go low, back track to
the switch & test again for a high state that goes low. If it works at the
Switch but not at the chip at 9C then look for a break in the trace between the
two points. If it does not work at the switch then try using a jumper wire
attached the the PCB ground (same as the logic probe ground point) and quickly
touch 9C pin 9. If all works now, suspect your switch. If not, suspect the
74LS161 @ 9C. You can hold your logic probe to 9C pin 9 and touch the ground
jumper wire to the logic probe tip if you want to observe the state change.
(joystick, coins, buttons, dip switches)
Note the 74LS367 has 2 sections with 2 seperate enables !
You can use the same technique as above to troubleshoot input switches such as the joystick, start buttons, Coin inputs or even the Dip Switches. You will notice that all of these switches have pull-up resistors. A pull up resistor brings the voltage up to +5 VDC so that the inputs are at a normally High state (1). On Pac-Man the pull-up resistors are RM6 (Dips), RM7 (test/table/start/player 2 controls), and RM8 (coin, credit, player 1 controls). The RM is a designation for Resistor Network or Resistor Array. You will notice above that RM7 is described as a 1K x 8 (1 K by 8), meaning that this resistor array has 8 resistors of 1K ohm value with a common lead.
You will also see a small .1 micro Farad capacitor (C17) tied to ground, this is to lengthen the momentary pulse (a low state as the switch is closed connecting it to Ground) as the capacitor recharges and enable the logic to actually have enough time to register a input. This also acts as a "debounce circuit" eliminating the circuit from registering many switch closures and opens that occur in the tiny time window as the switch changes state. (Notice the Dip Switches don't use these capacitors).
You can "see" the controls/ dips at work with your logic probe from the switch. Place your logic probe at the 74LS367 (Hex Buffer Tri-State) 8H pin 6. It will be normally High until you press the player 1 START down, when it will change to Low. So if I look at the output of this circuit, I'll see the same thing at 8H pin 7 right? WRONG. Notice the output is labeled DB5 (Data Bus 5), but DB5 is also at 8E pin 7, 8D pin 5, 3F pin 6, 4D pin 7, and 6D pin 6. We're on the BUS here man! Not only is this now the path for information about the joystick being pressed down but it could also be a Player 1 Down, or Dip 5 or info to the Graphic ROM's or Sound Section etc... The BUS is the pathway for data to travel in either direction. In our case the Data BUS is 8 BITs wide, our joystick closure is only one bit of information in a system that can directly access 64K bits.
So how do we keep up with this information? The same way the post office
delivers your mail, you have a address that is explicitly defined to you. Okay,
what's the address for "Player 1 Start"? We'll save that for latter. For now
it is important to notice that if you are on the bus, things will be going too
fast to see what you want. If you notice the 74LS367 chip at 8H has a line at
pin 1 IN1 which is an input (see the arrow pointing to our
chip?). This is used to ADDRESS our chip 8H and tell it "Hey man, what data do
you have for me on DB4~DB7" (it also reads DB0~ DB3 from 8F concurrently) at
which time the Central Processor Unit (CPU) is only looking at these particular
inputs at this given address, which includes our switch Player 1 START. So
things are too fast for us to know if they are working properly on the BUS side
but we can check to see if they are working properly before the Hex Buffer and
we can also be sly & check to see that the Hex Buffer is being used by the CPU
if we look at the IN1 signal at pin 1, it must go LOW to enable
a read from chip 8H.
The Clock Circuit is the primary time keeper for the circuit board, much as
a drummer is for a band. Pac-Man uses a single crystal to provide the timing
for the Z80 and develop the Video synchronization signals for the monitor
output.
The crystal (precisely cut piece of quartz crystal) vibrates at the given frequency and produces a tiny voltage swing. This voltage swing is captured between a pair of inverter gates of a 74LS368 at 8B. Output through one more inverter gate & used to drive a TTL input of the 74LS107 @ 8A.
The logic probe will not test a crystal properly (overload it's output) so
check the output of 8B pin 7 for a pulse. Note the enable lines of the 74LS368
@ 8B must have pins 1 & 15 low at all times to work. If you do not have an
output at 8B pin 7 suspect the crystal.
Okay, let's find a pulse...... Upper left corner of your schematics for Pac-Man is the CLOCK circuit. You will see 3 signals listed here.....
6 MHZ
6 M*
6 M*
Probe chip 8B pin 13. All three LED's should light and your probe should purr at a fairly high speed. (It is important that both High & Low LED's light, if one is missing this could indicate a chip that is not fulfilling the requirements of TTL logic and may cause problems down the line, spent many a hour troubleshooting a Xenon pinball board because I overlooked this....)
If you are adventuresome you can follow the 6MHZ signal through the
divider circuit at 8C, 3R, 3S. Here the signal is divided down to produce the
VIDEO SYNC signals for the monitor. Follow the 16H, 32H, 64H and 256H
to 2P, 3P and eventually out to 5N pin 11 where we have CMPSYNC.
'The logic probe can find many problems very quickly & it doesn't take long for you to mem(orize) chip outputs,e.g., quite common in older boards are the 74161 series that output at 11/12/13/14 and the popular 74157 series that output at 4/7/9/12. I always check outputs & if I find one bad, I drop back to the inputs & if they are ok...time to change the chip.
You can detect more complex troubles after getting used to using a logic
probe & knowing what to expect, e.g., 74161 outputs starting at pin 11 &
listening to the pulse tone & looking at the LED flashes, you learn to expect a
doubling of that when moving to pin 12, and again to pin 13, and once more to
pin 14. If at some point, you fail to hear this or it sounds the same on 2
adjacent pins, you know to further investigate this chip." Bob Roberts
So far we know how to check for RESET, in order for our game to run we
know this line must be high. If the line is pulsing then we have a condition
where the "watchdog is barking". Watchdog circuit acts like any good watchdog,
it observes the PCB (Pac-Man) in action by monitoring a select few ADDRESS Lines
for activity. Lack of activity within a preset number of machine cycles will
alert the watchdog that something is wrong. Like all good watchdogs, he "barks"
or send a signal out on the RESET line (74LS161 @ 9C pin 15) that forces the CPU
to restart over as if the power had been turned off.
What makes the watchdog bark?
Basically the CPU is hung in a loop, confused for some reason be it
Our logic probe will not help at finding these faults unless they are of the Open Circuit, Short Circuit or Stuck Circuit variety that we can see with the probe. So that is what we check for along these lines.
BUT FIRST, we must be sure that we have all of the correct timing signals that the watchdog timer uses to count down to a "bark".
Assuming that you have worked through the Sync generation section above (8C, 3R, 3S, 2P, 5M, 2R, 2S) with good results. Next check VBLANK for a pulse at the 74LS161 @ 9C pin 2. Missing a pulsing VBLANK ? Go to the 74LS10 at 3P, pins 8, 9, 10, 11 then to the 74LS74 at 5M pins 2, 3, 5, 6 (be sure pins 1 & 4 are pulled high).
The 74LS161 will generate the "bark" or watchdog reset signal that will send
a pulse to the RESET line then on through the 74LS02 @ 7L where it becomes the
RESET line to the Z80 pin 26.
Also know as Logic Tables. These list the input & output conditions for a gate
or circuit. If you picked up a TTL Cookbook or the like you will have examples
of these. Depending on your drive to learn all the inner workings of digital
logic you may decide to memorize all the gates and their functions. All are
math based (AND, OR, NOR, INVERTING etc...) so it will be fairly easy once you
associate the symbol with it's name.
Truth tables can also be constructed for larger circuits or in our case the entire Pac-Man board. Done on a chip by chip, pin by pin level at given conditions, say intermission or Game Over Demo. Each pin will have a listing of.....
P.....yellow pulse like xtal
LP....low pulse
HP....high pulse
HLP...high to low pulsing
L-above for a situation where an output (in attract) is low
for 20 to 30 seconds and then does one of the above & recycles.
H-& above
L....low
H....high
Bob Roberts suggested these and it is a great idea (now if someone could just
create the entire series and post it here for us!).
Now you have a way to tell if signal is valid or atleast cycling through states. This will not pinpoint many problems, example error in the output of the divider circuit for the video sync signal, the signal may cycle from both states & sound appropriate as far as you can tell but it would still be wrong. Heck the entire game may work if you adjust your monitor sync a bit. This would not be easy to find with a O-scope.
A logic pulser may seem like a wise addition to your troubleshooting tools. Also called a signal injector, it appears similar to a logic probe and is used in conjunction with a logic probe. A logic pulsers' purpose is to generate a pulse (usually 5 Hz or so) to input into a circuit that you are monitoring with your logic probe. 5Hz is a relatively low rate that is easy to see with the probe. It is a bit tidier and safer than a jumper wire to ground as we used early in troubleshooting SW1.
Many techs do most of their troubleshooting with a logic probe, it is quick &
easy. It helps find open circuits, short circuits & stuck states. It is a good
start
.
"This is far & away, the cheapest & easiest way for a collector to T-shoot a
board, once the basic skills are captured." The Real Bob Roberts™
