PINBALL: Repair DataEast/Sega 1987-1995 Pinball Games, part three

Support this Pinball Repair Website & PHoF. Please purchase the Marvin3m.com/top

Repairing DataEast 1987 (Laser War) to Sega 1995 (Batman Forever), Part Three
by cfh@provide.net, 06/10/08.
Copyright 2000-2008, all rights reserved.

Scope.
This document is a repair guide for DataEast and Sega pinball games made from 1987 (Laser War) to 1995 (Batman Forever). Sega games are included from 1994 to 1995 as Sega bought DataEast in 1994.

Internet Availability of this Document.
Updates of this document are available for no cost at http://marvin3m.com/fix.htm if Internet access is available. This document is part three of three (part one is here, and part two is here).

IMPORTANT: Before you Start!
IF YOU HAVE NO EXPERIENCE IN CIRCUIT BOARD REPAIR, YOU SHOULD NOT TRY TO FIX YOUR OWN PINBALL GAME! Before you start any pinball circuit board repair, review the document at http://marvin3m.com/begin, which goes over the basics of circuit board repair. Since these pinball repair documents have been available, repair facilities are reporting a dramatic increase in the number of ruined ("hacked") circuit boards sent in for repair. Most repair facilities will NOT repair your circuit board after it has been unsuccessfully repaired ("hacked").

If you aren't up to repairing pinball circuit boards yourself or need pinball parts or just want to buy a restored game, I recommend seeing the suggested parts & repair sources web page.

Table of Contents

1. Getting Started:

2. Before Turning the Game On:

3. When Things Don't Work:

4. Finishing Up:

3j. When things don't work: Lamp Matrix

Lamp matrix of 8 rows and 8 columns.

The lamp columns are controlled by TIP42 transistors that switch the 18 volts on and off many times within a second. The lamp rows are controlled by TIP122/102 transistors that switch the ground on and off. Because the TIP42's source the current (instead of sinking the ground like a TIP122/102), lamp column TIP42 transistors go bad more often than TIP122/102 lamp row transistors.

The most common CPU Controlled Lamp Problem.
The most common controlled lamp problem occurs when the fuse blows (with the blue/white wires) that connects to the large 30,000 mfd capacitor and the bridge in the backbox. If this happens, ALL the CPU controlled lamps will not work! Check these three components for problems which would cause this fuse to fail.

Overly Bright Lamps.
When a transistor or diode goes bad, generally it shorts on. When a transistor shorts on in the lamp matrix, it will make all the lamps in that row or column appear permanently on, and be very bright. This happens because the lamp matrix is actually 18 volts that is continually turned on and off, a row or column at a time. This nets a lower +6 volts that the lamps require. The lamps are never allowed to get full brightness at 18 volts before being switched off. If a transistor has shorted on, a row or column of lamps will be turned on for a longer time, and hence be brighter.

All the computer controlled lamps in the lamp matrix should flash in attract mode, or in the "All" lamps diagnostic test. If a number of lamps remain on (and they aren't general illumination lamps), there may be a lamp matrix transistor problem.

If a number of lamps are out, check the bulbs and fuses first. If a number of lamps are stuck on, check the game manual and see if they are in the same row or column. If so, test the individual transistor (see the Testing Transistors and Coils section) before replacing it.

Left: 44/47 lamp, socket and the orientation of the diode.
Note the banded end of the diode goes to the "middle" lamp lug.
The non-banded end goes to the lamp's top lug.
Right: The playfield socket used for 555 lamps. The small metal
tabs on the outside of the socket often get bent. This prevents
a good connection to the board on which they plug. Bend them
back for better contact.

Left: the component (lamp) side of a lamp board. Note the diodes mounted to the
board, and the use of 555 bulbs.
Right: the solder (socket) side of a lamp board. Note the Molex header pins soldered
here. Often these Molex pin solder joints crack or become fatigued, preventing the
lamp(s) from working.

Testing a diode on a lamp socket circuit
board. The black lead is on the banded
side of the diode.

Lamp Socket Diodes.

If a lamp socket diode has broken (become open), or is disconnected from the lamp socket, its lamp will not light.

Two Lamps On Instead of One.
If a lamp diode is shorted on (or installed incorrectly), this can cause two lamps to act as one. This can be seen in the "Single" (or "One") lamp test. Each individual lamp in the lamp matrix (as displayed on the screen) should flash. The "start game" button will move the test from one lamp to another. If TWO lamps flash in this test instead of just one, suspect the chosen lamp has a bad or mis-installed lamp diode.

Testing a Lamp Socket Diode.

Testing the lamp matrix columns using two test leads, a 555 socket
(pulled temporarily from the playfield), and a 1N4004 diode. One test
lead is attached to pin 1 of CN6 on the CPU board, and is stationary.
The other end is attached to the lamp socket. Another test lead is
connected to the second lead of the lamp socket. A diode is clamped
to the other end of the test lead. Then the non-banded side
of the diode is touched to each pin of connector CN7. In the
"all" lamps test the lamp should flash for each pin.

Testing the Lamp Columns.
If a TIP42 transistor that drives a lamp column is suspected as bad, test it:

Remove the backglass.
Turn the game on.
After the game boots, for games Frankenstein and before, press the green button inside the coin door into the down position. Then press the black button to enter diagnostics. Press the black button to move through the different tests. For portals games (Baywatch and Batman Forever), press the black diagnostic button, and use the flipper buttons to go to the diagnostics icon. Then press the game's start button.
Go to the "All" lamps diagnostic test.
Unplug the connectors at CN6 and CN7 (lower middle portion of the CPU board).
Connect an alligator test lead to pin 1 of CN6. Pin 1 is the right most pin, as facing the board.
Connect the other end of this test lead to one lead of a 555 light socket. Temporarily get one of these from a playfield lamp (make sure it has a working bulb first).
Connect another test lead to the second lead of the 555 light socket.
On the other end of the test lead, clip on a 1N4004 diode, with the non-banded end away from the alligator lead. Touch the non-banded end of the diode to pin 1 of CN7 (the columns plug). Again, pin 1 is the right most pin, as facing the board.
The lamp should flash.
Move the diode/alligator lead on CN7 to the next pin. Again, the lamp should flash.
Repeat the previous step, until the last pin of CN7 is reached.

If a lamp column tested doesn't give a flashing test lamp, that column is bad (or the test diode is reversed!). No light or a non-flashing, bright lamp are signs that the respective TIP42 column transistor is bad. Test the transistor as described in Testing Transistors and Coils.

Testing the lamp matrix rows using two test leads, a 555 socket
(pulled temporarily from the playfield), and a 1N4004 diode. One test
lead is attached to pin 1 of CN7 on the CPU board, and is stationary.
The other end is attached to the light socket. Another test lead is
connected to the second lead of the lamp socket. A diode is clamped
to the other end of the test lead. Then the banded side of
the diode is touched to each pin of connector CN6. In the "all" lamps
test the lamp should flash for each pin.

Testing the Lamp Rows.
If a TIP122/102 transistor that drives a lamp row is suspected as bad, test it:

Remove the backglass.
Turn the game on.
After the game boots, for games Frankenstein and before, press the green button inside the coin door into the down position. Then press the black button to enter diagnostics. Press the black button to move through the different tests. For portals games (Baywatch and Batman Forever), press the black diagnostic button, and use the flipper buttons to go to the diagnostics icon. Then press the game's start button.
Go to the "All" lamps diagnostic test.
Unplug the connectors at CN6 and CN7 (lower middle portion of the CPU board).
Connect the alligator test lead to pin 1 of CN7. Pin 1 is the right most pin, as facing the board.
Connect the other end of this test lead to one lead of a 555 light socket. One can be temporarily borrowed from a playfield lamp (make sure the lamp works first!).
Connect another test lead to the second lead of the 555 light socket.
On the other end of the test lead, clip on a 1N4004 diode, with the banded end away from the alligator lead. Touch the banded end of the diode to pin 1 of CN6. Again, pin 1 is the right most pin, as facing the board.
The lamp should flash.
Move the diode/alligator lead on CN6 to the next pin. Again, the lamp should flash.
Repeat the previous step, until the last pin of CN6 is reached.

If a lamp row tested doesn't give a flashing test lamp, that row is bad (or has the test diode reversed!). No light or a non-flashing, bright lamp are signs that the respective row TIP122/102 transistor is bad. Test the transistor as described in Testing Transistors and Coils.

Testing the Lamp Rows (returns); Alternative Method.
The lamp rows are controlled by TIP122/102 transistor at Q72 to Q79. With the game in diagnostics test mode (menu or any test but the lamp tests), run an alligator jumper wire from ground to the metal tab on one of these TIP122/102 transistors. This will cause the entire lamp row to light up.

Testing the Lamp Matrix Columns and Rows with a Logic Probe.
If a logic probe is available, it can be used to easily test the lamp matrix:

Remove the backglass.
Turn the game on.
Game can be in "attract" mode, or in the Test menu's "ALL" lamp test.
Unplug the connectors at CN6 and CN7 (bottom portion of the CPU board).
With the logic probe connected to power and ground, probe each pin 1 to pin 9 of CN6 (pin 1 is the right most pin, as facing the board). These are the lamp rows (return). All pins should show PULSE (low) on the logic probe.
With the logic probe connected to power and ground, probe each pin 1 to pin 9 of CN7 (pin 1 is the left most pin, as facing the board). These are the lamp columns (drive). All pins should show PULSE on the logic probe.

The Lamp Matrix Power Resistors.
On the CPU board at the lower right corner, there are eight 0.4 ohm, 3 watt wire wound sand resistors (R55 to R62). Sometimes these resistors get hot and de-solder themselves from the CPU board and fall off (or the solder joints need to be re-soldered). If there is a lamp matrix problem and it's not the row or column transistors, make sure you check these resistors.

Most Common Problems with Lamps.

Bad bulb. Any light bulb can burn out. Often the bad bulb can be seen, but sometimes not. Test the bulb with a DMM, set to continuity. Put the test leads on the bulb. No continuity, and the bulb is bad.
Wire broken away from the socket. This happens quite often and requires re-soldering the wire back to the socket lug.
Diode broken away from the socket. If the lamp diode becomes disconnected from its socket, the lamp will not light.
Corroded or Bad Socket. Games imported back into the US from other countries exhibit this problem more often. Re-seating the bulb in its socket often fixes this problem. On 555 plug-in sockets, bend the contact tabs slightly for better contact.
Blown Fuse. If many GI lights don't work, check the fuse associated with them. If all the CPU controlled lamps don't work, check the F2 fuse by the big 30,000 mfd capacitor bolted to the back of the backbox.
Burned Connector on the power supply or interconnect board. This happens most often with GI (general illumination) lamps. See Burnt GI Connectors for more info.
Lamp matrix power resistors. On the CPU board, lower right corner, there are eight 0.4 ohm, 3 watt wire wound sand resistors (R55 to R62). Sometimes these resistors get hot and de-solder themselves from the CPU board and fall off (or the solder joints need re-soldering).
Bad Column Transistor. The TIP42 transistors that control the lamp matrix columns often fail. If this is the case, all the lamps in a particular column will be brightly lit all the time.
Bad row diode on CPU board. Easy to check with a DMM set to diode function.
Two Lamps act as One. If a lamp diode has a shorted on, this can cause two different lamps to act as one.

I replaced the Offending TIP122 and TIP42, and I've check the Lamp Matrix Power Resistors, but the Lamp Matrix still doesn't Work Right.
It is rare, but problems in the lamp matrix can occur further up stream.

For the lamp matrix columns, transistors Q56 to Q63 are 2N6427 transistors which drive the TIP42 column transistors. Sometimes these can fail too. Before the 2N6427 transistors are two 7408 TTL chips at 9F and 10F which drive the whole circuit. All these components connect to the 6821 PIA at 11D. Any of these things could fail.

For the lamp matrix rows, 7406 TTL chips at 11F and 12E drive the rows though a series of 2N5060 (NTE5400) silicon controlled rectifiers (SCR) at Q81 to Q88. Also there is a diode on the CPU board connected to each Row SCR, check this with a DMM as sometimes these go bad. These all connect to the 6821 PIA at 11D.

Chase Lamps (Rope Lights) used on Star Trek 25th and Hook.
The chase lamps (rope lights) used on the ramp on ST:25th and Hook are controlled by a little board in the upper right corner of the backbox. This board has a 74HCT04 and a ULN2003. The "running lamps" diagnostic test strobes the inputs to this board, the 74hct04 inverts the input signals, and the outputs should be pulled to ground by the ULN2003. A logic probe is sufficient for testing this. Most often the ULN2003 goes bad on this board.

Testing Star Trek 25th Ann. rope lights with a 9 volt battery.

The lamp strings used on the ramps consist of three sets of four lamps. The molex connector going to the rope lights has pins for [+10VDC] [Key] [Drive1] [Drive2] [Drive3]. The rope lights can be bench tested with a 9V battery and couple of alligator test lead. Put +9 volts on the input pin, and touching the other (negative) battery lead to one of the three drive lines. Four lights on the ramp chase string will come on if it's working correctly.

3k. When things don't work: Switch Matrix

If a switch is not activated in several games, or is permanently closed, the switch is assumed to be bad. This will create a test report, which is shown when the game is turned on (this feature was implemented in DataEast games Phantom of the Opera and later). If a particular feature of a game is difficult to score, it's associated switch may be (falsely) assumed bad (if not activated in so many games). To correct the test report, remove the playfield glass, and activate the switch by hand within a game, or within the diagnostics switch edge test.

A switch report, shown when the game is turned on.

Switch Matrix "Flow" (Logic Path).

The switch columns (drive) are controlled through CPU board connector CN8, which flows like this:

CN8 switch column connector on CPU board.
RA8 resistor network (1k x 8, very common failure point and difficult to test without removing it from the board).
Q48-Q55 transistors 2N3904. Another very common switch failure point.
R98-R105 resistors 1.5k.
RA11 resistor network (4.7k x 8).
8J chip 74LS244.
8H PIA 6821 pins 10-17.

The switch rows (return) are controlled through CPU board connector CN10, which flow like this:

CN10 switch row connector on CPU board.
RA16 resistor network (560 ohm x 8).
R116-R133 resistors 1k. NOTE schematic shows R116-R133 as two resistor networks RB5 and RB6 (1k x4), but implemented on most CPU boards as discrete resistors.
D26-D33 diodes 1N4004.
5J/6J chips 4011.
8H PIA 6821 pins 2-9.

Diagnosing Common Switch Failures.
If several switches don't work on a game, first make sure it's not a playfield wire problem. For example, maybe a switch wire has broken "upstream", as the 8x8 rows/column wires are daisy changed from switch-to-switch, making all switches "downstream" not work.

Using the switch matrix table in the game manual, see if the non-working switch(es) are all part of a particular row or column in the switch matrix. To do this you'll need the game manual, and to have the game in Switch Edge test. Depressing switches on the playfield should show the correct switch number on the score display. If one switch shows multiple switch closures, or shows the wrong switch closure, or shows nothing, you have a switch matrix problem! Again try and see if the problem lies all in the same switch matrix row or column. A shorted switch diode on the top side of the playfield (from a ball crashing into the switch diode, bending it and shorting the switch leads) can cause very wacky switch behavior. So always do a top-side playfield examination if you have a switch problem.

Aside from that, the next most common problem is caused by user error. That is, shorting the coil voltage (be it 30 or 50 volts) to a switch. This will for sure blow the corresponding 2n3904 switch matrix column transistor on the CPU board at Q48-Q55, and maybe do much more damage than that. This problem can often be diagnosed using the Switch Edge test, and am entire switch column does not work at all, or signals switches in other columns.

The key here is to determine if the switch problem is located on the CPU board, or on the playfield. This way you'll know where to start diagnosing the problem. Put the game in the Switch Edge test and check all the switches for problems. If a problem is found, with the game still in Switch Edge test, CPU board connectors CN8 and CN10 can be removed. Then using an alligator test lead and a 1n4004 diode, the CPU board can be tested directly without the playfield connected (this is explained below). If this test checks out Ok, then the problem is on the playfield. This is important information to have otherwise you could be chasing a switch problem down the wrong path.

Switch diagnostics (Baywatch and later) have become
very informative. This is the switch test known as "TST".
Shown is the switch wiring, CPU connectors, wire colors
and even the switch's column (drive) transistor (Q49).

Shorting the Switch Matrix to +50 volts.

After replacing the suspected CPU board components, disconnect the switch input plugs from connectors CN8 and CN10 at the bottom of the CPU board. Put the game into switch test mode, and none of the switches should be activated on the display! If a whole row of switches is still activated, this would indicate something in the CPU circuitry is still bad.

Shorting the Flipper EOS switch to the Lane Change switch.
Many games have a feature called "lane change". This allows the player to change the lit lanes using the cabinet flipper buttons. Prior to Robocop, this is done by doubling up a second switch on the flipper EOS switches. The EOS switches are a +50 volt switch. The lane change switch is a +5 logic switch connected to the switch matrix. These two switches are insulated from each other by a small nylon "triangle" activator. But if these two switches touch, the switch matrix will burn (as described above).

Games Monday Night Football and before:
the EOS flipper switch on these games is
actually two switches. The switch on the
left (with the diode) is low-voltage and connected
to the switch matrix. This normally open
switch closes when the flipper is activated,
and controls the lane change and high score
initials. The switch on the right is the high-voltage
flipper EOS switch. A nylon space insulates the
two switches from each other.

When adjusting or cleaning the flipper EOS switches or lane change switches, make sure the game is turned OFF. This will prevent shorting these two switches together. Also, do not clean the smaller lane change switch with anything other than a business card.

Lane Change on games with Solidstate Flipper boards.
When the solidstate flipper system was introduced on Robocop, DataEast stopped using a doubled up switch on the flipper EOS switches for the lane change. The lane change switch was moved to the cabinet flipper switch. This was done because the cabinet flipper switch on solidstate flipper games was a low voltage switch. This meant there was less potential switch matrix problems with the switch located here instead.

Test Button Switches.
The test button switches inside the coin door are wired directly to the CPU board at connector CN14. These switches do not go through the switch matrix. The test button switches are direct switches to the CPU board. The diagnostic switch connects to the CPU's NMI line. DataEast did this so when there is a switch matrix short, the user can still get into the game's diagnostic tests without using any switches in the switch maxtrix.

Switch Numbering.
Each switch has a number associated with it. The switches are just numbered 1 to 64. For example, switches 1 to 8 are column 1, rows 1 to 8. Likewise, switches 9 to 16 are column 2, rows 1 to 8. This continues up to switches 57 to 64, which are column 8, rows 1 to 8. (This is unlike Williams' WPC games, where switch 42 means column 4, row 2.)

Using the Internal Switch Tests.
To test switches, use the internal test software. On Frankenstein and before, press the green button inside the coin door down, then press the black button. Pressing the black button again will move from test to test. Go to the switch level ("TST") test. On Baywatch and Batman Forever, just press the black diagnostic button inside the coin door to enter the service portal. Use the flipper buttons to scroll to the "diag" icon, and the game's start button to select it. Then go to the switch level "TST" test. Activate any switch on the playfield using a pinball (this simulates real game play), and it should show on the game's display.

If a Bad Switch is Found.
If a switch does not work, check these things:

On the top side of the playfield, make sure the bad switch contacts haven't been bashed together by an errant "air" pinball.
On micro-switches, check the actuator arm. Make sure it's adjusted properly. Listen for the micro-switch's "click" when activating. No click usually means the switch is mis-adjusted or broken.
Check that the wires going to the switch are soldered well, and haven't fallen off.
Check the continuity (using the DMM's continuity setting) of the daisy-chained wire between this switch and another working switch in the same column (green wire) or row (white wire).
If it's a blade or leaf style switch, check the contacts for proper closure. Clean the switch contacts with a business card (do NOT use a file as the contacts are gold plated). Put the card between the contacts, close the contacts, and pull the card through the contacts once or twice. This is all that is needed to clean gold plated switch contacts.
Check the switch to make sure it works. Use the DMM's continuity setting, and put one lead on the "common" lug (the lug to which the banded end of the diode connects) of the switch. Put the other lead on the green (normally open) switch lug. The meter should only beep when the switch is activated, and not beep when the switch is de-activated. Move the DMM's lead from the green to the white wire (normally closed) switch lug. The meter should beep when the switch is de-activated, and NOT beep when activated.
Check the diode on the switch. Make sure the diode is connected properly, and is working (see "bad switch diode" or "fail safe diode test" below).
Check other switches in that switch's row or column. Two 4041 chips control rows, and 2N3904 transistors and 74LS244 at 8J control columns. Both rows and columns are controlled by a 6821 PIA at 8H.

If the switch is bad, replace it. If all the switches are bad in a particular switch column or row, chances are good the problem is not the switches themselves (CPU board problems).

Switch matrix of 64 switches: eight rows and eight columns.

Switches are "Daisy Chained".

Phantom Switch Closures: a Shorted Switch.
It's a strange problem. During game play, the ball goes down the right inlane, and the left slingshot fires! Or when a ramp shot is made, the game slam tilts. One switch closes, but a completely unrelated event then occurs.

This is a classic description of a shorted switch. It confuses the switch matrix into thinking something else has occurred. This can happen from an "air" pinball, that bashes an above playfield switch's contacts together, causing a short. Also a bad switch diode can do this. In either case, the shorted switch must be found. Unfortunately, it won't be obvious. The switch matrix is confused, so any diagnostics the game provides will be of limited help.

First, try to find the switch that causes something unrelated ("phantom") occurance. Take the playfield glass off, and start a game. Activate the switches with your hand, and find the switch which activates the phantom (unrelated) switch. Once the switch is found, go to the game manual and find the switch's number, row number, and column number. Say for example, switch 53 (column 7, row 5) is causing the phantom closure. Now get the other three switches that make up the remaining corners of the "square" of the row and columns. First get the reverse switch number, switch 39 (column 5, row 7). Then get the other two "corner" switches: switch 37 (column 5, row 5), and switch 55 (column 7, row 7). The switch short will probably be one of these four switches.

If there are problems figuring out if the short is in the playfield or the CPU board, try this. Remove connectors CN10 and CN8 from the CPU board. Next put the game in switch edge test. Using the manual, find which row and column of the switch that causes the phantom closure. Then cross this row and column directly on the CPU board (with wire and alligator clips, and a diode, as described below in the "Testing the Switch Columns/Rows"). The row and column numbers for each pin of connectors CN10 and CN8 are listed below. If the phantom switch does not activate, the problem is in the playfield. If the phantom closure still works, there is a CPU board problem.

If the phantom switch problem is on the CPU board, don't forget to look at the 1k ohm x 8 resistor pack RA6 on the CPU board. When this resistor pack goes bad, it can cause intermittent phantom switch closures. Use the DMM, and test the resistor pack. If in doubt, just replace it.

Bad Switch Diode.
Each micro-switch on the playfield also has an 1N4004 diode soldered to it. This diode can short closed. It doesn't happen often though.

Important: If a switch diode does short closed, all switches in that particular column or row will exhibit strange behavior. If a switch diode goes permanently open, the switch will never register. Keep this in mind when diagnosing switch matrix problems.

Fail-Safe Diode Test.
With the game off, use the DMM set to diode position. With the black lead on the banded side of the diode, it should show a .4 to .6 volt reading. Reversing the leads should show a null reading. If the diode tests bad, cut one lead of the diode from the switch to remove it from the circuit, and test it again just to make sure the diode is really bad. Replace the diode if bad, or reconnect the diode after testing.

Testing a switch diode on a microswitch without removing the
diode. The switch must be activated, and the middle green wire
(ground) must be disconnected.

Testing a Microswitch's Diode, without removal.

Disconnect the middle green (ground) wire from the switch. It should have a quick connector. If the middle green ground wire is soldered to the switch, ignore this test and do the above "fail-safe" diode test.
Put the DMM on diode setting.
Connect the black lead of the DMM to the diode's banded side, and the red lead to the non-banded side.
Activate the switch.
A reading of .4 to .6 should be seen.
Reverse the DMM's leads (red lead to the diode's banded side), and keep the switch activated. A null meter reading should be shown.

Testing a Blade/Leaf Switch's Diode.
Testing the diode on a leaf switch is far easier. No wires need to be disconnected, and the switch should not be activated. This technique assumes the switch is wired in the standard configuration: green (ground) wire to the center lug, the banded end of the diode solo, and the non-banded diode lead and the switch wire(s) to the other switch lug (as shown in the pictures below).

Leave the leaf switch's diode and all wires connected.
Make sure the switch isn't activated.
Put the DMM on diode setting.
Connect the black lead of the DMM to the diode's banded side, and the red lead to the non-banded side.
A reading of .4 to .6 should be seen.
Reverse the DMM's leads (red lead to the diode's banded side). A null meter reading should be seen.

Installing a New Switch Diode.
Replace the diode with a 1N4004 (or 1N4002 or 1N4001) diode. Make sure to install the new diode with its band in the same orientation as the old diode (assuming it was correctly installed in the first place, which isn't always a good assumption). If unsure, compare the diode's band orientation to a working switch and diode. Most (but not all!) switches have the green (ground) leads connected to the center (normally open) lead of the switch. Then the row (white) wire is connected to the switch lead closest to the center lead (the normally closed lead). The banded end of the diode is connected solo to the far (common) switch leg, and the non-banded end is connected to the same leg as the row (white) wire. There are some exceptions to this mounting. The game manual should specify any non-standard switch installations.

Notice the orientation of the diode's band on these switches. On a micro-switch,
the ground (green) wire usually goes to the center lug, the "live" wire and
the non-banded side of the diode to the lug closest to the center. The band
on the diode goes to the solo, far third switch lug. The leaf switch uses the same
connection method (ground to center, banded end of diode solo). Note there are
some exceptions to this mounting.

Switch Matrix Row or Column Problems: the Easy Test.
If the game is giving an error message regarding a switch matrix row or column problem, determine if this is a CPU board problem or a playfield problem. The easiest way to do this is to unplug the switch matrix row and column plugs at CN10 and CN8. Enter the game's diagnostics (coin door green button down, press the black button), and go to the switch edge test. If the row or column problem is gone (no switch reports), the problem is in the playfield wiring. If the problem is still there, the problem is on the CPU board. Be aware that most shorted switch matrix problems are caused by a bad switch matrix column 2N3904 transistor at Q48-Q55 (which affects the entire column).

Switch Matrix Plug and Pin Numbers.
When doing intensive switch matrix diagnostics with the plugs removed at CN10 and CN8, simulate an actual playfield switch closure, without using the playfield! This can be done by using the internal switch edge test, and an alligator lead connected to the particular row/column (switch) in question, and a diode (as described above).

CN8 Switch Column (drive) Pin Numbers

Pin 1 = Column 1, green/brown
Pin 2 = Column 2, green/red
Pin 3 = Column 3, green/orange
Pin 4 = Column 4, green/yellow
Pin 5 = Column 5, green/black
Pin 6 = key
Pin 7 = Column 6, green/blue
Pin 8 = Column 7, green/violet
Pin 9 = Column 8, green/gray

CN10 Switch Row (return) Pin Numbers

Pin 1 = Row 8, white/gray
Pin 2 = Row 7, white/violet
Pin 3 = Row 6, white/blue
Pin 4 = key
Pin 5 = Row 5, white/green
Pin 6 = Row 4, white/yellow
Pin 7 = Row 3, white/orange
Pin 8 = Row 2, white/red
Pin 9 = Row 1, white/brown

Testing the switch matrix columns: Using a diode and a test lead,
the test lead is attached to pin 9 of CN10, and is stationary. The
other clip holds the non-banded side of the diode. Then the banded
side of the diode is touched to each pin of connector CN8. The
"switch levels" test should indicate switches 1, 9, 17, 25, 33, 41, 49, 57
when moved from pin 1 to 9, respectively.

Testing the Switch Columns (drive).

Remove the backglass.
Turn the game on.
After the game boots, go to the diagnostic menu's switch levels or "TST" test.
Unplug the connectors at CN8 and CN10 (bottom portion of the CPU board).
Connect an alligator test lead to pin 9 of CN10. Pin 9 is the left most pin, as facing the board.
On the other end of the alligator test lead, clip on a 1N4004 diode, with the banded end away from the alligator lead. Touch the banded end of the diode to pin 1 of CN8 (pin 1 is the right most pin, as facing the board).
The display should show switch 1 is closed.
Move the diode/alligator lead on CN8 to the next pin. The display should show switch 9 is closed.
Repeat the previous step, until pin 9 of CN8 is reached. Switches 1, 9, 17, 25, 33, 41, 49, 57 should be closed on the display as the lead is moved forward, pin 1 to pin 9, on connector CN8. Note pin 6 is a key pin, and should be skipped.

If a particular column number does not display as closed, or is closed without any test lead connection, there is a problem on the CPU board. Usually this is a bad switch matrix column 2N3904 transistor at Q48-Q55.

Testing the switch matrix rows: Using a diode and a test lead,
the test lead is attached to pin 1 of CN8, and is stationary. The
other clip holds the banded side of the diode. Then the non-banded
side of the diode is touched to each pin of connector CN10. The
"switch levels" test should indicate switches 8, 7, 6, 5, 4, 3, 2, 1
when activated.

Testing the Switch Rows (return).

Remove the backglass.
Turn the game on.
After the game boots, go to the diagnostic menu's switch levels or "TST" test.
Unplug the connectors at CN8 and CN10 (bottom portion of the CPU board).
Connect an alligator test lead to pin 1 of CN8. Pin 1 is the right most pin, as facing the board.
On the other end of the alligator test lead, clip on a 1N4004 diode, with the banded end towards the alligator lead. Touch the non-banded end of the diode to pin 1 of CN10 (pin 1 is the right most pin, as facing the board).
The display should show switch 8 is closed.
Move the diode/alligator lead on CN10 to the next pin. The display should show switch 7 is closed.
Repeat the previous step, until pin 9 of CN10 is reached. Switches 8, 7, 6, 5, 4, 3, 2, 1 should be closed on the display as the lead is moved forward, pin 1 to pin 9, on connector CN10. Note pin 4 is a key pin, and should be skipped.

If a particular row does not display as closed, or is closed without any test lead connection, there is a problem with the CPU board.

Testing the Switch Matrix Columns and Rows with a Logic Probe.
If a logic probe is available, it can be used to easily test the switch matrix:

Remove the backglass.
Turn the game on.
Game can be in "attract" mode, or in the Test menu's "Switch Levels" test.
Unplug the connectors at CN8 and CN10 (bottom portion of the CPU board).
With the logic probe connected to power and ground, probe each pin 1 to pin 9 of CN8 (pin 1 is the right most pin, as facing the board). These are the switch columns. All pins should show PULSE on the logic probe.
With the logic probe connected to power and ground, probe each pin 1 to pin 9 of CN10 (pin 1 is the left most pin, as facing the board). These are the switch rows. All pins should show HIGH on the logic probe.

A short between two lamp matrix transistor metal tabs. Pic by J.Gerold.

Two Lamp Matrix Columns Shorted.

Resistor Network Testing/Explaination used in the Switch Matrix.
The resistor networks in the switch matrix can fail, and this is fairly common. There are two types of resistor networks used on the DataEast/Sega CPU board; "Bussed" and "Isolated".

If a resistor network is "560 x 8" and has 9 pins, that means it's BUSSED; all the resistors are tied to one common pin (this pin is labeled with a white square around it on the circuit board). Bussed resistor networks are labled as "RAx" (where "x" is a number). Simply put, if a resistor network has an odd number of pins, it is probably bussed.

If a resistor network is "1K x 4" and has 8 pins, this is an ISOLATED resistor network; a bunch of resistors in the same small package, so it uses two pins per resistor. Isolated resistor networks are labeled as "RBx" (where "x" is a number). Simply put, if a resistor network has an even number of pins, it is probably isolated.

Left: A bussed resistor network. Right: An isolated resistor network.

Here are the different resistor networks used in DataEast/Sega switch matrix:

RA16: 560 ohms x 8, bussed, used for the switch returns (rows).
RB5/RB6: 1k ohms x 4, isolated, used for the switch returns (rows).
RA6: 1k ohms x 8, bussed, used for the switch drives (columns).
RA11: 4.7k ohms x 8, bussed, used for the switch drives (columns).

When testing a BUSSED resistor network, first find pin 1. This pin will have a white square around it, to isolated it from the rest of the pins. Use a DMM set to ohms, and put one lead on pin 1. Put the other lead on each pin 2 to 9. The same reading should be seen for each pin 2 to 9.

When testing an ISOLATED resistor network, put the DMM leads on the two adjacent pins furthest to the right or left, and note the reading. Then move both DMM leads down one pin. The same value should be seen. Continue down the resistor, moving both DMM leads one pin at a time, until all adjacent pins are tested.

A bussed resistor network on a DataEast game. Note the white
square around pin 1; this is the common pin.

A "hacked" bussed resistor network (same DataEast game as above).
If the correct resistor network can not be found, this shows how to
make a bussed resistor network from individual resistors. Note the
white square around pin 1; this is the common pin. This picture shows
quite well how a bussed resistor network is wired. Since some resistor
networks are hard to get, this repair will get more and more common.

www.digikey.com

members.aol.com/_ht_a/espain5941/myhomepage

1k x 4 issolated (4 pins): Q4102-ND (discontinued)
1k x 8 bussed (9 pins): Q9102-ND (discontinued)
4.7k x 8 bussed (9 pins): Q9472-ND
560 x 8 bussed (9 pins): Q9561-ND
3.3k x 8 bussed (9 pins): Q9332-ND (discontinued)

Unfortunately 9 pin resistor networks are not very popular anymore. Usually everybody uses an even number of pins (6, 8, 10) even if it gives you an odd number of resistors (5, 7, 9). The even pin numbered devices are far easier to find than the odd pinned ones. Digikey seems to have dropped all 9 pin parts in favor of 8 and 10 pin parts. If really needed, can always buy 10 pin parts and cut off the 10th pin. Bourns seems to be the only common manufacturer of these now. Most are available from mouser.com.

1K x 4 resistors Isolated (8 pins), Bourns part# 4608X-102-102 (Mouser part# 652-4608X-102-1K). Digikey carries a very similar Bourns part# 4308R-102-102 (Digikey part# 4308R-2-102-ND).
1K x 8 resistors Bussed (9 pins), Bourns part# 4609X-101-102 (Mouser part# 652-4609X-101-1K).
4.7K x 8 resistors Bussed (9 pins), Bourns part# 4609X-101-472 (Mouser part# 652-4609X-101-4.7K).
560 x 8 resistors Bussed (9 pins), Bourns part# 4609X-101-561 (Mouser doesn't have stock but lists it as part# 652-4609X-101-560 or 652-4310R-101-560). If needed a 10 pin version of this part can be used and any additional pin cut off (make sure the correct pin is cut!), Bourns part# 4610X-101-561 (Mouser part# 652-4610X-101-560). .
3.3K x 8 resistors Bussed (9 pins), Bourns part# 4609X-101-332 (Mouser part# 652-4609X-101-3.3K or 652-4310R-101-3.3K).

Bad Switch Column (Drive): How to Diagnose/Fix.
Usually the switch column transistors fail. These are 2N3904 transistors at Q48 to Q55. See the Checking Transistors section for details on testing these transistors. Right before the transistors there is a bussed resistor network at RA6 (1k x 8) which should be check (as described above). After the resistor network is a serious of small capacitor, C74 to C81 (these rarely fail). Next each one of the eight 2N3904 transistors connects to a 1.5k resistor at R98 to R105. Then the individual resistors connect to a bussed resistor network at RA11 (4.7k x 8). This network can fail. A reading of 4.7k ohms for each connection should be shown:

CN8 pin 1, to Q55, to R105, to RA11 pin 2, to 8J pin 18 (column 1).
CN8 pin 2, to Q54, to R104, to RA11 pin 3, to 8J pin 3 (column 2).
CN8 pin 3, to Q53, to R103, to RA11 pin 4, to 8J pin 16 (column 3).
CN8 pin 4, to Q52, to R102, to RA11 pin 5, to 8J pin 5 (column 4).
CN8 pin 5, to Q51, to R101, to RA11 pin 7, to 8J pin 14 (column 5).
CN8 pin 6: KEY
CN8 pin 7, to Q50, to R100, to RA11 pin 8, to 8J pin 7 (column 6).
CN8 pin 8, to Q49, to R99, to RA11 pin 9, to 8J pin 12 (column 7).
CN8 pin 9, to Q48, to R98, to RA11 pin 10, to 8J pin 9 (column 8).

Bad Switch Row (Return): How to Diagnose/Fix.
First check the 560 ohm X 8 bussed resistor network at RA16 (as described above). Then check the two isolated 1k ohm x 4 resistor networks at RB5 and RB6 (as described above). A reading of 1k ohms should be shown for each connection:

CN10 pin 1, to RA16 pin 2, to RB5 pin 1, to RB5 pin 2 (1000 ohms), to 6J pins 1 & 2 (switch row 8).
CN10 pin 2, to RA16 pin 3, to RB5 pin 3, to RB5 pin 4 (1000 ohms), to 6J pins 12 & 13 (switch row 7).
CN10 pin 3, to RA16 pin 4, to RB5 pin 4, to RB5 pin 6 (1000 ohms), to 6J pins 8 & 9 (switch row 6).
CN10 pin 4: KEY
CN10 pin 5, to RA16 pin 5, to RB5 pin 7, to RB5 pin 8 (1000 ohms), to 6J pins 5 & 6 (switch row 5).
CN10 pin 6, to RA16 pin 6, to RB6 pin 1, to RB6 pin 2 (1000 ohms), to 5J pins 1 & 2 (switch row 4).
CN10 pin 7, to RA16 pin 7, to RB6 pin 3, to RB6 pin 4 (1000 ohms), to 5J pins 12 & 13 (switch row 3).
CN10 pin 8, to RA16 pin 8, to RB6 pin 5, to RB6 pin 6 (1000 ohms), to 5J pins 8 & 9 (switch row 2).
CN10 pin 9, to RA16 pin 9, to RB6 pin 7, to RB6 pin 8 (1000 ohms), to 5J pins 5 & 6 (switch row 1).

Jurassic Park Switch Matrix Problems.
On the game Jurassic Park, one of the main playfield wire harness trunks can rub against the upper right flipper coil stop. The vibration of the flipper coil can cause the harness to saw against the coil stop, shorting the wiring. This can cause random coil firing, blown fuses, and switch matrix shorts. To fix this, first inspect the wires for any cuts or shorts. Then resecure the wiring harness with nylon wire ties. Use the ties to pull the wiring harness away from the coil stop. This problem was discussed in service bulletin number 53. View this bulletin by clicking here and here.

Hook Erratic Slam Tilt Problems.
Intermittently Hook performs strangely, and occassionally "slam tilts" or kicks solenoids for no apparent reason. This can be caused by a intermittently grounded switch strobe line. Most often, this is caused by the green wire with the yellow trace on the left drop target bank. This wire gets pinched against the playfield support bracket. To fix this, check this wire on the drop target, and make sure it is not pinched by the support bracket. Dress the wire with electrical tape or heat shrink tubbing, and a tie wrap, to prevent future occurances. This problem was discussed in service bulletin number 37.

Further Diagnosing of the Switch Matrix.
If there is a switch matrix problem, the first plan of attack is to do the above column and row switch matrix tests. If these tests pass, the problem most likely is in the wiring. Note most switch failures show as row failures (even though it could be a column problem). Here are several different ways the switch matrix can fail. All require the use of the internal "switch level" or "switch edge" or "TST" tests of the game.

Switch column shorted to ground.
When a column wire is shorted to ground, and any switch in that column is closed, the switch test will show ALL switches in the ROW of the closed switch as being closed. If no switches are closed, the switch test will show no switches closed. To find the location of the short, go to the end of the switch column wire on the playfield (the switches are "daisy chained" together for an entire column or row). Then break the daisy chain one switch at a time until the short no longer shows in the switch test.
Row shorted to ground (diode anode).
When the anode (non-banded end of the switch diode) is shorted to ground, the switch test will show the entire row as activated (whether any switches are closed or not). To find the location of the short, go to the end of the switch row wire on the playfield (the switches are "daisy chained" together for an entire column or row). Then break the daisy chain one switch at a time until the short no longer shows in the switch test.
Row shorted to ground (diode cathode).
When the cathode (banded end of the switch diode) is shorted to ground, that switch's entire row will show as closed in the switch test (whether the switch is open or closed). To find the location of the short, go to the end of the switch row wire on the playfield (the switches are "daisy chained" together for an entire column or row). Then break the daisy chain one switch at a time until the short no longer shows in the switch test.
Column wires shorted together.
When two column wires are shorted together, and none of the switches in those columns are closed, the switch test will show no problems. But pressing any switch in either column will show that switch, along with a switch in the column that is shorted on the row of the switch being closed. For example, if column 2 and column 4 are shorted together, closing switch column 2 row 3 will also show a closed switch in column 4 row 3.
Row wires shorted together.
When two row wires are shorted together, and no switches are closed, the switch test will show no closed switches. When any switch in either row is closed, another switch on the same column as the closed switch will also show as closed. For example, if rows 1 and 4 are shorted, closing a switch in row 1 column 3 will also show a closed switch on row 4 column 3.
Column and row wires shorted together.
When a column and row wire are shorted together, the switch test will show the switch that is at the intersection of the row and column as being closed, even though it is not closed. All other switches on all other rows and columns will work correctly. For example, column 1 and row 3 are shorted together. The intersection of this column and row will show that switch as closed (even though it's not). And remember, this switch is not the cause of the problem!
Open diode on a switch.
An open diode on a switch will cause only that switch to not work.
Shorted diode on a switch.
A shorted switch diode will show no problems when only that switch is opened or closed. However if additional switches in that row or other columns are closed, false switch readings can be shown.

Switch Maintenance.
Here are the procedures for maintaining the switches:

Micro-switch: no maintenance required. Can adjust the actuator arm only by rotating the switch in its bracket. Do not BEND the activator arm! Loosen the two screws holding the switch, and rotate the switch to adjust the activator arm. Re-tighten the screws, but not too tight as it will bind the switch mechanism.
Blade or Leaf switch: clean with a business card inserted between the contacts. Squeeze the contacts closed, and remove the business card. Do not use a file on these gold plated contacts! Re-adjust the contact spacing for correct operation.
Opto switches: use a Q-Tip and some Windex. Dip the Q-tip in the Windex, and clean the opto's two LED's (receiver and transmitter) with the Q-tip.

3L. When things don't work: Ball Trough Problems & Optic Switches

Opto Overview.
Sega designed their optic system to operate just like a standard switch. Blocking the light beam caused the normally open opto switch to respond by closing the switch, just like a mechanical switch (and unlike Williams' optic switches, which are normally closed). Sega used ultra-bright visible red light LEDs (Light Emitting Diodes), so an opto transmitter's light can be seen with the naked eye (again, unlike Williams which used infrared light that cannot be seen with the naked eye). These ultra bright optos have a narrower bandwidth too, so ambient light has little affect on them (again, unlike infrared LEDs). Because of this, troubleshooting is easy:

With the game on, enter diagnostics, switch test.
Block the optic light beam; does it repond in the test? If yes, the optic switch is OK. If no, continue...
Visually inspect the optic transmitter; is it glowing red? If yes, than the transmitter is OK. If no, check pin 2 of the opto transmitter board for +5 volts. Remember the opto transmitter is doing nothing more than acting like a flashlight. If it is on, it is working!
If the opto switch is still not working, remove the 2 pin connector with the green and white wires on the opto receiver board (the board that does NOT have the glowing LEDs). Using a jumper wire, short the two wires on this connector plug together. This should show the switch as closed in the switch test. If the switch does close with the short, there may be a bad alignment of the opto receiver and transmitter (this is a very common problem). If shorting these two wires together does nothing, suspect a broken wire in the switch matrix wiring.
Also try using a flashlight and shine it into the opto receiver. To do this, be sure to totally block the red transmitting opto. Does the flashlight change the status of the optic switch as the light is moved in front of the optic receiver? If so, and the opto transmitter is working (shining red), there is an opto alignment problem.
Another idea: If the receiving opto stays closed as shown in the switch test (even if the light source is blocked or not), double check the opto alignment. Also clean the lenses of the optos. A good trick is to remove both the receiving and transmitting optos, and move them very close together to test them. Does the switch status change in the switch test? If so there is an alignment or dirt problem with the transmitter and receiver optos, which may require some tweaking to get the optic switch to work correctly.

Ball Trough Optics.
Starting with Baywatch, Sega added a single LED (light emitting diode) and opto receiver above position one of the ball trough (to sense that the ball has been kicked to the playfield shooter lane). This morphed into a two LED ball trough system a bit later.

Replacement Optics.
The opto transmitters are the part that most often fails. The specs needed are:

Size: 1 3/4 (T size) or 5mm.
Illuminated Color: Red
Lens Color: Clear
Wavelength: 644 nm ("Ultra Red" type)
View Angle: 10 to 12
mA (max): 20
Volts (typ): 1.7
Volts (max): 2.5
MCD (typ): 3000

These are "Ultra Red" LEDs, which are in the 644 nm range. Not that "Super Red" (630 nm) don't really work. There is a VERY sharp cutoff in the wavelength range. The Opto used by Sega and up to Stern Terminator3 was MT5000UR, made by Marktech Opto. These are now obsolete and no longer available. Next Stern started using the Toshiba TLRH180P in the late 1990s (for example, Sharkey's Shootout and Austin Powers in 2000), and these are still used in their games today. But the TLRH180P was dropped due to RoHS issues, and replaced by TLRH180P(F), witch is the RoHS compliant version. There is a substitute by Fairchild MV8114, with a frequency of 660 nm. Doesn't seem like much compared to 644 nm, but it's enough to make them barely work together. (For this reason, replace them in pairs.) Though a "Super Red" (not "Ultra Red") LED, it has a sloppy center frequency, which is wide and high in range. For this reason the MV8114 just barely works. But then Fairchild discontinued their entire visible LED product line and handed it over to Everlight (Chinese), which still sells the MV8114. Note I have used the MV8114 in games from Maverick to Simpsons, and it works just fine (as long as you replace both the emitter and receiver, to get a "matched pair").

Interestingly the MT5000UR and TLRH180P are used for *both* the emitter and receiver. This isn't done often, but it does work, and it's what Sega/Stern is doing (note the receiver and transmitter are both the same part number). When the MT5000UR was used as a receiver, it had a 4.7k ohm resistor going to the Drain "D" leg of the 2n5460 Jfet amplifier and Base "B" leg of the 2n3906 transistor. When Stern changed to a TLRH180P, this resistor changed to a 5.6k ohm type. If replacing either the emitting LED or detecting LED, I suggest you always change them both together to keep the wavelengths matched.

How does this work, using an LED as both the transmitter and receiver? As Ed describes, the LED is not acting as a phototransistor, but more as a photodiode. Much the same way that a diode can be forced to emit light when current is flown through it, the LED can also conduct current when sufficient light is forced into it. When LEDs are used to detect light as a photodiode, they have a lower 'gain' than say a phototransistors. A phototransistor can (barely) provide sufficient current loads with light into it. (I say "barely" because this is seen in the Williams' WPC pinballs, where they need a special QVE11233.0086 to drive high current load). A photodioide can provide barely any current with the same quantity of light. So it requires an amplifier on the output (the 2n5460 Jfet and 2n3906) as seen on the Stern receiver boards. This is how Stern gets around using special high-current optical switches in their designs; they jump amplify the output.

A Major Opto Board Problem.
Problems with the opto's themselves are rare. A much more common problem is cold/fatiqued solder joints on the small PCB which holds the optic circuit. The LED's are mounted to two small 2"x2" boards, with several surface mounted transistors. Because the ball trough takes a lot of abuse from the balls falling into the trough, it is very common to see cold or fatiqued solder joints on ball trough optic boards. Often the solder joints on the connector head pins and the surface mounted parts fatique (crack), from the constant jarring of the balls in the trough. This can cause intermittent (or complete) failure of ball trough sensing. To fix this problem, use a low heat solding pencil, and reflow the solder on these surface mounted transistors. Though the components are small, this can be easily done. Also reflow the solder joints on the header connector pins on these small boards.

The surface mount components on the ball trough
optic board. These tiny components can get fatigued
(cracked) solder joints from balls constantly
slamming into the ball trough.

Opto Service Bulletin.

here

Later Whitestar system Sega games use TWO optics in the ball trough. The second set of optos is used to verify a ball is actually plunged to the ball shooter lane. This system still has the same fatiqued/cracked solder joint problem on the optic's PC board.

Roller Switch Ball Troughs (Jurassic Park and later).
Starting with Jurassic Park, all DataEast/Sega games used a five or six ball trough with a roller micro switch for each ball. These switches can get damaged very easily from balls dropping on them. If any one of these switches does not work properly, the game will not work correctly. Error reports such as "ball missing" or multiple balls served to the plunger lane are common problems. Luckily, these switches are easily replaced with a common Radio Shack part number 275-017 ($1.99).

Often the problem with these roller micro switches is the last one or two balls to enter the ball trough will not activate their respective switch. If the switch is not moving freely, or the switch's roller not working perfectly, the ball may not roll onto the switch (activating it). If the switch cannot be adjusted to fix this, replace it. Do not try to lubricate the switch! Just replace the switch.

Testing the Ball Trough Roller Switches.
Most of the games with the 6 ball trough roller switches have a feature in the diagnostics called "Easy Trough Clear". Upon entering this diagnostic feature, pressing a flipper button will shoot one ball from the trough into the shooter lane. The dot matrix display should show how many balls the game thinks are left in the trough. By doing this, one ball at a time, a defective switch or balls that don't roll down the trough are pretty easy to identify.

After all the balls are cleared from the trough, go to the switch test. Make sure each roller switch activites in the switch test. If more than one switch is shown to activate with a single roller switch, there is some sort of switch matrix short or a mis-wired switch. This can confuse the game and cause problems. See the switch matrix section of this document for more details.

The Ball Trough Roller Switch Diodes.
Remember there is a diode on each ball trough roller switch. It is possible for these diodes to get broken, crushed or shorted when the playfield is lowered from a raised service position. Keep this in mind too.

"Six Balls Missing" Error on JP.
Another problem seen, for example on Jurassic Park, is the message "six balls missing". Clearly not all six ball trough switches are broken! Instead, transistor Q54 (on Jurassic Park) has failed. The lock ball assembly coil (mounted under the ball arch on the top side of the playfield) has its red wire (32 volts) located very close to the mounting screw for the ball trough assembly. This creates a potential short, causing transistor Q54 to fail. To prevent this, make sure the solder lug on the coil is insulated from the ball trough mounting screw using electrical tape or heat shink tubing. This problem was discussed in service bulletin number 53. View this bulletin by clicking here and here.

Three Ball Troughs (Laser War to Rocky & Bullwinkle).
Prior to the roller switches, DataEast games (Rocky & Bullwinkle and prior) used large micro switches with long metal activator arms. This ball trough system was limited to three balls maximum. Problems with bent or broken ball activator arms are common on this ball trough system.

A five ball trough with the five roller switches.

3m. When things don't work: Score Display Problems

The "normal" DMD (128x32) is by far the least expensive of the three displays (about $125 street price). The small (132x16) display sells for about $165, and the super sized (192x64) display sells for over $200 (that's why DataEast/Sega discontinued it's use).

How is the DMD implemented on DE pinballs?
The dot matrix display is run by a separate computer. This is unlike say Williams who have a single 6809B that does all the DMD animations AND runs the lamp and switch matrix and coils. On DE games, there is a 6809 on the big CPU board that runs the lamp and switch matrix and coils, and then a separate CPU that runs just the DMD animation. This separate DMD CPU board is mounted directly behind the display itself, and is either a Z80, 6809B, or 68000 (depending on the size of the dot matrix display screen).

The two computers communicate via a ribbon cable. When the main CPU board needs an animation, it does a "call command" to the DMD CPU, which does the animation. You can sometimes see this on games that have mis-matched ROM software. For example if the main CPU is running ROM version 2.x and the DMD CPU is running ROM version 1.x, sometimes the DMD will show "Command 24" on the screen instead of the actually animation. This happens because the animation call number does not exist in the older DMD ROM version. For this reason the main CPU and the DMD CPU should both have matching version numbers (say 2.x for both the CPU and the DMD ROMs). In fact if the ROM versions are mis-matches, some pretty wacky score display screen behavior can be seen.

In addition to the communications between the main CPU and the DMD CPU via the ribbon cable, the PIA 9b on the main CPU board can cause problems too. This PIA (peripheral interface adaptor), if damaged, can cause bad "calls" to the DMD computer. This can also show wacky DMD screen behavior and strange "command call" errors (like "command 9c") on the score display.

Display Service Bulletins.
For more information on dot matrix score displays, check out DataEast's service bulletin number 50 by clicking here and here. For more information on alpha-numeric score display (as used on Laser War to Simpsons), check out Sega's service bulletin number 87. It explains the entire theory of operation for these displays. For this service bulletin click here and here.

Dot Matrix Display works for 5 minutes than goes blank.
This is a very common problem with all DataEast dot matrix displays, and is often related to the connector bring power to the display's computer board (mounted on the back dot matrix panel), or power to the display itself. On the large sized DMDs especially, the 5 volt .156" Molex connector going to the DMD's computer often needs new Trifurcon pins crimped into place (red and black wires usually). Also the 5/12/HV power going from the power supply to the display itself often develops bad .156" connectors pins too (on the display and/or the power supply), requiring new terminal pins.

"Outgassed" Score Displays.
Score display glass has a limited life; it does not last forever. Time will eventually kill these, and the display will "outgas" and fail. Because of the high voltage involved with score displays, the anode and/or cathode inside the display glass breaks down. This results in the "outgassing" of impurities that eventually change the internal gas properties, so the display can't glow (the gas must be very pure for the display to work). There is no fix for this other than replacing the display glass (which has become fairly expensive now). Outgassing affects dot matrix displays very often, but it also applies to alpha-numeric displays too (see the alpha-numeric section below for more details).

An outgassed score display puts much higher electrical loads on the power supply. When a display starts to outgas, it should be replaced immediately! Otherwise, the repair may involve fixing a damaged (burnt) power supply, in addition to replacing the score display glass.

The blue circle shows a dot matrix display that is outgassed.

Dot Matrix Displays (DMD).

The Dot Matrix Controller Board's CPU and PAL Chips.
DataEast/Sega games Checkpoint and later with a dot matrix display have a dot matrix controller board. This board has separate CPU and EPROM chips which only handle the dot matrix score display animations. This CPU is in addition to the CPU board's 6808 (or 6802) CPU chip, which controls the lamps, solenoids, switches, games rules, etc.

Checkpoint, Batman, Ninja Turtles, Star Trek 25th, Hook (128x16 dot display): Z80A CPU. This dot matrix controller is mounted right to the back of the dot matrix score display glass itself.
Lethal Weapon to Guns N Roses (128x32 dot display): 68B09 CPU on a separate controller board mounted behind the display glass, #520-5055-00 (Lethal Weapon to Last Action Hero) or #520-5055-01 (Tales from the Crypt to Guns N Roses).
Maverick to Batman Forever (192x64 dot display): 68000 CPU on a separate controller board mounted behind the display glass, #520-5092-01.

The 128x16 and 128x32 dot matrix display games use a very different power supply than the previous alpha-numeric power supply. Instead of just -100/+100 volts and +5 volts, the dot matrix displays need -98, -110, +68, +5 volts (and sometimes +12 volts).

The 128x32 dot matrix controller #520-5055-00 and 520-5055-01 uses two custom PAL chips at U2 and U16. Both of these chips are very static sensitive, and can fail if the board is mis-handled (U16 seems to be more problematic of the two). The PAL code is available here for download, if a PAL programmer is available. For location U2, a TIBPAL16R4-25 chip is needed (code checksum 50D8). For location U16, a TIBPAL16L8-15 chip is used (code checksum 6489).

EPROM size for Medium size DMD Controller Board.
On the medium (128x32) DMD controller board, there are two different versions #520-5055-00 (Lethal Weapon to Last Action Hero) and #520-5055-01 (Tales from the Crypt to Guns N Roses). The only difference is the size of the EPROM(s) the board is jumpered to use. The -00 board has resistor R11 installed, and is configured for 27020 sized EPROMs. The newer -01 board has resistor R11 removed which configures it for the larger 27040 EPROMs. So if you want to use the larger 27040 EPROMs, make sure resistor R11 is removed from the DMD controller board.

Voltages Needed for the DMD Controller Board.
Since all the dot matrix displays essentially have their own computer on the controller board to control the score display, a supply of good +5 volts is needed for the controller. If a display is not working, first check the controller board for +5 volts. Often the connector carrying +5 volts can gain resistance, which will lower the +5 volts. This can be enough to make the DMD not work (this problem happens most with the super-sized 192x64 DMD).

The display itself is a complete separate unit, and does not get voltage from the display controller. On the super-size 192x64 DMD, the input voltages are low (+5 volts and +12/18 volts coming into the display). Sega did this because the 192x64 DMD has it's own power supply on the dot matrix controller board itself (the higher voltages are not supplied by the main power supply). This is unlike the other two smaller dot matrix displays.

Below are the Power Supply connector CN5 output voltages.

Alpha-Numeric power supply connector CN5:

CN5 pin 1: ground
CN5 pin 2: NC
CN5 pin 3: -100 volts
CN5 pin 4: +100 volts
CN5 pin 5: key
CN5 pin 6: +5 volts

Alpha-Numeric System 11 power supply connector J5 (for comparison reference):

J5 pin 1: ground
J5 pin 2: NC
J5 pin 3: -100 volts
J5 pin 4: +100 volts
J5 pin 5: key
J5 pin 6: +5 volts

Small 128x16 DMD power supply connector CN5:

CN5 pin 1: ground
CN5 pin 2: -98 volts
CN5 pin 3: -110 volts
CN5 pin 4: +68 volts
CN5 pin 5: +12 volts
CN5 pin 6: +5 volts
CN5 pin 7: key

Medium 128x32 DMD power supply connector CN5:

CN5 pin 1: ground
CN5 pin 2: -98 volts
CN5 pin 3: -110 volts
CN5 pin 4: +68 volts
CN5 pin 5: +12 volts
CN5 pin 6: +5 volts
CN5 pin 7: key

Large 192x64 DMD power supply connector CN5. No high voltage feed to this board. The HV is generated on the dot display board itself.

CN5 pin 1: ground
CN5 pin 2: ground
CN5 pin 3: ground
CN5 pin 4: +5 volts
CN5 pin 5: +5 volts
CN5 pin 6: +5 volts
CN5 pin 7: key
CN5 pin 8: +5 volts

Listed below are the other voltages required for the three different DMD displays, and the pin out. All voltages should be within 10% plus or minus of the numbers listed below:

Small 128x16 DMD, connector J2:

J2 pin 1: +5 volts
J2 pin 2: Ground (logic)
J2 pin 3: KEY
J2 pin 4: +12 volts
J2 pin 5: +68 volts
J2 pin 6: -98 volts
J2 pin 7: -110 volts
J2 pin 8: Ground (high voltage)
J2 pin 9: Ground (high voltage)

Medium 128x32 DMD, connector P1:

P1 pin 1: -110 volts
P1 pin 2: -98 volts
P1 pin 3: KEY
P1 pin 4: Ground
P1 pin 5: Ground
P1 pin 6: +5 volts
P1 pin 7: +12 volts (only used on Babcock displays)
P1 pin 8: +68 volts

DALE Super-Size 192x64 DMD, connector P3:

P3 pin 1: +5 volts DC
P3 pin 2: Ground
P3 pin 3: Ground
P3 pin 4: +24 volts DC. (Dale schematics show 24 volts, but 16-18 volts is probably used, as the voltage originates from a backbox mounted bridge rectifier BR1 or BR3 supplying 18 volts DC.)
DALE (only) component P1 voltages. Derived by the on-board power supply. The voltages are taken from the above connector P3 and through a small power supply, converted to the following:
- P1 pin 8: +80 dc
- P1 pin 2: -90 dc (vrw)
- P1 pin 1: -100 dc
- P1 pin 4: Ground
- P1 pin 5: Ground

Babcock Super-Size 192x64 DMD, connector P3:

NOT INSTALLED.
Babcock (only) component P1 voltages. These are converted on the board to the +80, -90 and -100 volts DC needed for the display.:
- P1 pin 1: +12 dc
- P1 pin 2,3: Ground
- P1 pin 4: No connect

Cherry Super-Size 192x64 DMD, connector P3:

NOT INSTALLED.
Cherry (only) component P2 voltages. These are converted on the board to the +80, -90 and -100 volts DC needed for the display.
- P2 pin 1,2: +12 dc
- P2 pin 4,5: Ground
- P2 pin 6: +5 dc

WARNING: on small and medium dot matrix displays, the power plug on the glass display board itself is the *same* type of plug. But do NOT connect a small dot display to a medium display power source (or vice versa)! It will RUIN the display if this happens, as the pinout is *not* the same!

Small DMD Problems and Fixes.

The display on "Mutant Ninja Turtles" with a bad R95 resistor.

The same display after R95 is replaced.

R95 is replaced here on this small DE display board (Dale).

Another flavor of small DE display board, showing R95 (Cherry).

Vertical Lines in the Small DMD.
Another problem seen on the small DE dot matrix display is vertical lines, which completely distort the images behind it.

The vertical line problem in small DMD displays.

Testing the Small Dot Matrix Display Glass.
Use this technique to test a new or used small (128x16) dot matrix score glass. This technique uses an existing small dot matrix display panel to test a loose dot matrix score glass. Here is the procedure:

Turn the game off.
Lower the display panel on the game.
Connect a (black) alligator jumper wire from either lead of capacitor C29 (near U15) on the back of the display glass' circuit board.
Connect the other lead of the (black) alligator jumper wire to one of the row pins on the dot matrix glass. This connector puts -100 volts DC on the row the jumper is connected to. NOTE: there are two sets of row pins, on each side of the display glass. Each set corresponds to that half of the display glass.
Connect a (red) alligator jumper to the banded side (cathode) of CR2 on the dot matrix display board (this is a connection to +68 volts DC).
Put a 33k ohm 1/2 watt resistor in the free lead of the red alligator clip.
Turn the game on.
Touch the free end of the resistor (which is connected to the red alligator lead) to one column lead on the display glass. Note the column must be on the same half of the glass as the row connected!
The dot which intersects the row and column should glow.

number 76

Medium Sized DMD Problems and Fixes.

128x16 and 128x32 Dot Matrix Display "Failures".

If the backbox 18 volt lamp matrix fuse (8 amp slo-blo, with the blue/white wire connecting to the fuse holder) fails, this can cause the dot matrix display to "crash", causing these problems. Another way to identify this problem is the lack of any playfield CPU controlled lighting. The game will "play" (that is, the game will start and play balls), but with score display problems and no CPU controlled lamps. Simply replacing this the 18 volt lamp matrix fuse (in the backbox) will fix this problem (providing the associated components such as the bridge and capacitor are OK). If there is a lamp matrix power short, or the lamp matrix backbox bridge is bad, the problem will need to be fixed or the fuse will continue to immediately fail.

This problem occurs because the +12 volts needed for the dot matrix display is generated by the 18 volt lamp matrix bridge and the capacitor/fuse bolted inside the backbox (through connector CN5 on the power supply). The 12 volts generated by the power supply board (for the sound board which comes from connector CN6), does not provide this voltage to the dot matrix display! Hence the power supply could be working perfectly, and this problem could still exist.

A Babcock 128x32 display where the 18 volt lamp matrix backbox
fuse has failed. Notice the horizontal "garbage" line across the
bottom center of the display. This problem only seems to affect
the Babcock brand dot matrix displays.

This stange problem was solved with the advent of the 192x64 super-sized dot matrix display (Maverick to Batman Forever). An additional backbox 18 volt fuse (F3), bridge and capacitor was added. This backbox voltage supplied power to a switching power supply, implemented on the 192x64 DMD display driver board. This way if the lamp matrix (F2) fuse failed, the dot matrix display remained unaffected.

128x32 DataEast Dot Matrix Display EPROM differences.
Depending on the age of the game, the dot matrix display may have either a single 27040 EPROM, or two 27020 EPROMs. If it is necessary to change the number of EPROMs used in the 128x32 dot matrix display, add or remove resistor R11 on the Display controller board (the board that holds the display EPROMs). Here is the breakdown:

Two 27020 EPROMs = R11 installed (display board part number 520-5055-00).
One 27040 EPROM = R11 removed (display board part number 520-5055-01).

number 38B

Summary of Small & Medium sized DataEast/Sega DMD Fixes.

Check the voltages at the display board: -110, -98, 68, 12 and 5 volts (or within 10% of those values).
Double check the 5 volt connection to the display board.
Check the 18 volt backbox lamp matrix fuse.
Check data ribbon cable from CPU to display board.
On small 128x16 DataEast games, check/replace resistor R95 on the display.
Replace the ROMs on Display board.

Rebuilding the Small & Medium DMD HV Power Supply 520-5047-00,01,02.
On power supply #520-5047-00 (small DMD - Checkpoint, Turtle, Batman, Star Trek 25th), #520-5047-01 (Lethal Weapon, Star Wars, Rocky & Bull), and #520-5047-02 (Jurassic Park, Last Action, Crypt, Tommy, WWF, GnR) power supplies, often one or more of the high voltage power lines will be missing for the dot matrix display (-110, -98 or +68 volts). The power supply HV (high voltage) section will probably need to be rebuilt. The -00 power supply used a MJE340 and MJE350 transistor to generate the -110 and -98 volts. The -01 and -02 power supplies used two MJE15030 and a MJE15031 instead. That is essentially the differences between the three power supply revisions. A high voltage rebuild will require the following parts:

TR1 : MPS-A92
TR2 : MPS-A42
TR3 : MJE15030 or MJE340 (-00 originally used a MJE340 but either will work).
TR4 : MJE15031 or MJE350 (-00 originally used a MJE350 but either will work).
TR6 : MJE15030, -01 and -02 only (the -00 does not have a TR6).
D5,D7 : 1N4004 diode
D6,D8 : 1N4730 or 1N5228 (3.9 volt zener diode)
D9 : 1N4760 (68 volt zener diode)
D10* : 1N5379 (110 volt zener diode), -00 only. Note a 1N4764 (100v) or even a 1N4763 (91v) can be used in the -00, but the display won't be quite a bright.*
D10* : 1N4764 (100 volt zener diode), -01 Rev A,B.
D10* : 1N4743 (13 volt zener diode), -01 Rev C and -02.
D11* : 1N4742 (12 volt zener diode), -00 only.
D11* : 1N4743 (13 volt zener diode), -01 Rev A,B.
D11* : 1N4764 (100 volt zener diode), -01 Rev C and -02.
R15 : 500 ohm, 10 watt or 130 ohm, 5 watt (500 ohm used on -00, 130 ohm on -01 & -02).
R14 : 3.9k or 4k, 10 watt (-00)
R10,R11 : 330k, 1/2 watt (-00)
R12,R13 : 1.5k, 1/2 watt (-00)
R8,R9 : 47k, 1 watt flameproof (-00)
C12,C13 : .1 mfd 500 volt

* It's always best to check D10 and D11 before replacing. Use the same value as was originally installed.

On -00 power supplies (small DMD games), many imported games will replace the D10 diode (1N5379, 110v) with a 1N4004, which is incorrect. Using a 1N4004 at D10 on a -00 power supply will allow way too much voltage through and run the small DMD display much too bright. This was done on displays that are outgassed, to use them beyond their normal life, delaying the cost of replacing the display.

Note that power supply 520-5047-01 and 520-5047-02 are interchangable. But there are some differences in their circuits:

520-5047-01 Rev A,B
No resistor R17
D10: 1N4764 tied to ground
D11: 1N4743

520-5047-01 Rev C and 520-5047-02
Has resistor R17
D10: 1N4743
D11: 1N4764 tied to ground

High Voltage section on a small DMD power supply -00.

Large 192x64 Dot Matrix Display Problems and Fixes.

Also, when both flippers are simutaneously used during a game, the dot matrix display can reset, (but the CPU may not!) This will display the display board ROM revision level and the initial power-on sound board tones/speech.

The first step when dealing with a large DMD display "blanking out" is to replace the 5 volt power connector on the DMD. There is a three pin two wire (red/black) .156" Molex plug on the large DMD controller board. Re-pin this connector with new Molex Trifurcon .156" terminal pins. Often this fixes the blank-out problem.

Booting the Large DMD on the Bench.
The cool thing about the large DMD is how easy it is to boot on the work bench. Because the large DMD generates the high voltage right on the DMD board itself, it only needs 5 and 12 volts DC to boot. (Even though it says it needs 24 volts DC as an input voltage, these displays work fine with 12 volts DC instead.) Using a standard computer power supply, the 4pin hard drive connector can be plugged directly into the large DMD power connector! Using the same computer power supply, you can power a DE CPU board and DMD controller board with 5 volts. So with just a computer power supply, you can get a DE game (CPU board, DMD controller, large DMD display) up and running in attract mode right on the work bench.

Large DMD Power Cable Update Kit.
There is an update kit available from DataEast/Sega/Stern for the large dot matrix display games. This kit is a wiring harness that goes directly between power supply and the large dot matrix display which runs the +5 vdc directly from the power supply to the dot matrix display. The kit is part# 500-6326-00, and costs about $11. The kit can be purchased from Stern, or self-manufactured (many distributors do not stock this kit). Also the kit can be made for much less than $11.

If you want to make your own 192x64 upgrade kit, the parts needed are:

5 feet of #18 red wire.
5 feet of #18 black wire.
5 pin .156" molex plastic connector body.
8 pin .156" molex plastic connector body.
(4) .156" molex connector pin, crimp style.
Some heat shrink tubing.

Making and installing the 192x64 upgrade parts:

Remove the old connector from the Display control board connector J1. Remove the old connector on the Power supply board connector CN5, if present. These connector(s) will no longer be used.
Use five feet of 18 gauge red wire, create a "loop". Connect the two ends of the wire together, and solder them. Use some heat shrink tubing to insulate the wire jointing point. Repeat this with the 18 gauge black wire.
On one spot of the red wire loop, strip back the insulation about 1/4 inch without cutting the wire. Bend the wire in the middle of this stripped area. Crimp a .156" molex connector pin to this point. Carefully solder the wires to this connector pin after crimping. Repeat this on the wire loop 2.5 feet away (the two new pins are now directly opposite each other in the wire loop). Repeat this step for the black wire loop.
Insert the one red wire and one black wire connector pins into the 5 pin plastic connector body. One red wire pin (+5 volts) goes to connector pin 1. The black wire pin goes to connector pin 3. Note that pin 2 is the "keyed" pin. Use the Display controller board connector J1 as a reference for the pin numbers.
Insert the other red wire and black wire connector pins into the 8 pin plastic connector body. One red wire pin (+5 volts) goes to connector pin 8. The black wire pin goes to connector pin 2. Note that pin 7 is the "keyed" pin. Use the Power supply board connector CN5 as a reference for the pin numbers.
Attach the new 5 pin molex connector body to the Display controller board connector J1. Attach the new 8 pin molex connector body to the Power supply board connector CN5.

The above modification will give the large dot score display direct power, via brand new connectors, directly from the power supply. Note instead of using Power supply connector CN5, the present IDC connector at CN6 can also be used. The black wire can be inserted directly into the CN6 connector, pin 1. The red wire can be inserted directly into the CN6 connector, pin 10 (for reference, the key is pin 5).

View Sega's original service bulletin number 106 for this modification by clicking here and here.

Large DMD EPROM Speed.
Also be aware the large dot matrix display requires 120 nanosecond EPROMs or faster. If using 150 ns EPROMs (which are OK in the older 128x32 and earlier displays), the super size display will not work! The 68000 processor for the display is running at 12 mHz, so the 150 ns EPROMs are too slow.

If the Super Size DMD Still Doesn't Work...
Additionally, the +5/+12 volt bridge rectifier (DB1) on the power supply board may need to be replaced. If this bridge becomes "leaky" (less efficient), it may not be able to supply the robust +5 volts needed on the display board. So if the above modification does not work, replace bridge DB1 next. When replacing DB1 on the power supply board, also solder an 18 gauge wire from the "+" lead of bridge DB1 to the "+" lead of capacitor C4. Solder another 18 gauge wire from the "-" lead of the bridge DB1 (the lead diagonal to the bridge's "+" lead) to the "-" lead of capacitor C1. Do this on the solder side of the power supply board. These added wires will help prevent future cracked solder joints on the power supply board.

Remember there is also a BR3 bridge rectifier and a fuse mounted to the backbox metal ground plane. These provide voltage to the super sized score display. Sometimes this bridge can short or the lug connectors fail. Be sure to check that.

The backbox has an additional bridge rectifier
BR3 and large cap for the Super Sized DMD power.

Still doesn't work? Make sure the backbox grounding strap is tight and making good contact to the 192x64 dot matrix controller board. A loose grounding strap can contribute to super size DMD problems.

Last resort for 192x64 displays that are on again, off again, random dots, fading in and out, and rebooting. Stern suggests pulling and reseating the chips on the display board behind the DMD glass. The chips (three of them) may come out hard and may sound "gritty" on the way out and back in. But give it try.

Important Notes on Installing a New 192x64 Dot Matrix Display.
192x64 DMD score displays often need a small modification to work in the Sega games. This involves changing the supply voltage jumpers on the display itself. On older 192x64 displays, there is no jumper, but instead a trace is cut between the ribbon cable connector plug and the P3 power plug (see picture below).

An older 192x64 dot matrix display with the supply power trace cut to work in
Sega pinball games. Picture by Jerry.

On newer 192x64 DMD score displays there is a jumper marked "W1 Supply Pin". This jumper must be set correctly or the display will not work in Sega pinball games. The two vertical "4" jumper pads on the left should be tied together (see picture below).

A newer 192x64 dot matrix display with the supply power jumpers set to work in
Sega pinball games. Picture by Jerry.

Jurassic Park's Display Blanking Problem.
In game over (attract) mode, Jurassic Park's score display blanks out for about 15 to 30 seconds. In any coins are inserted, only the first coin will give a credit. This problem is a communications error between the CPU board and the Display board's CPU, which causes the CPU board to go into a wait state. This takes 15 to 30 seconds to time out! To fix this, install CPU EPROM software version 5.10. This problem was discussed in service bulletin number 53. View this bulletin by clicking here and here.

Changing the Games Default Language.
DataEast/Sega does not have DIP switches to select the default language used on their score displays (unlike Williams WPC games). The only way to change the language is to burn a new display EPROM (on games with a 192x64 display, remember to use 120 nanosecond or faster EPROMs). It is also recommended to update the CPU game EPROM at the same time, to the lastest version of the game's software. Stern has all the DataEast/Sega game's EPROM software available on their web site at http://www.sternpinball.com/rom.htm.

Alpha-Numeric Display Problems.

The 39k ohm Power Supply Resistors.
The major villain in the DataEast alpha-numeric power supply is resistors R8 and R9. These are both 39k ohm, 1 watt resistors. Very often either or both of these resistors will go out of spec, or even completely open. This will prevent the +100 and/or -100 volts from getting to the score displays.

If the alpha-numeric score display is barely working, one may think it's the display glass itself. This is in fact a common problem, where the displays "outgas" and don't show as brightly, and eventually completely die. But before changing the expensive score display glass, change the 39k ohm resistors at R8 and R9 on the power supply with new 1 or 2 watt "flameproof" resistors.

If your high voltage fuses are not blown, and your score displays do not work, first replace the two 39k ohm resistors. These are cheap and easy parts to replace (a lot cheaper and easier than replacing score display glass). At the very least check these. Replace these two resistors with "flame proof" 1 or 2 watt 39k ohm versions. Make sure the new resistors are mounted slightly off the circuit board, so air can get under them for cooling.

Check the +100/-100 volts at the Power Supply board.
It's easy to check the +100 and -100 volts at the power supply board. These voltages should be from 95 to 105 volts. Check for these voltage at power supply connector CN5.

CN5 pin 1: Ground
CN5 pin 2: Not used
CN5 pin 3: -100 volts DC
CN5 pin 4: +100 volts DC
CN5 pin 5: KEY
CN5 pin 6: +5 volts DC

Power Supply Diodes Leak.
Like the two 39k ohm resistors, the two diodes at D6 and D8 (1N5228) can start to "leak". If the displays are weak and the 100 volt section is low (even with the score display glass unplugged), try changing these two diodes (after changing the 39k ohm resistors first).

Blown high voltage score display Fuse F3.
Every DataEast game has a fuse to protect the +100 and -100 voltages which power the score displays. This fuse is F3 on the power supply board, a 1/4 amp 250 volt slo-blow fuse. If this fuse is blown, remove power supply connector CN5, replace the fuse, and power the game back on. If it blows again immediately, the high voltage section of the power supply board will probably need to be rebuilt.

Also a blown UDN7180 chip on the display controller board can cause the high voltage fuse(s) to blow on the power supply board. These chips can short the +/- 100 volts directly ground, and blow the fuse. If power supply fuse F3 only blows when power supply connector CN5 is attached, a shorted display glass or a failed UDN chip could be the problem.

Increasing Alpha-Numeric Score Display Life.
Glass score displays are getting very EXPENSIVE. Because of this, it is important to make the current glass score displays last as long as possible. The best way to do this is to decrease the 100 volts (which powers the displays) to 91 volts. This can be done by replacing the zener diodes at D9 and D10 with a lower voltage diode

The original diodes used at D9 and D10 are 1N4764. These are 100 volt, 1 watt zener diodes. If replaced with 1N4763 diodes, which are 91 volt, 1 watt zener diodes, only 91 volts (instead of 100 volts) will power the displays. This will make the displays slightly less bright, but it will also DRAMATICALLY increase their life span! Since glass score display tubes are becoming so expensive, this is highly recommended.

Rebuilding the Alpha-Numeric 100 volt Power Supply Section.
If any of the high voltage fuses are blowing, the 100 volt power supply section will probably need to be rebuilt. The following parts on the power supply board should be replaced.

Positive Alpha-Numeric 100 volt section parts to replace:

TR3 = MJE340 or MJE15030 transistor. The MJE340 is a NPN 300 volt transistor, and slightly more robust than the MJE15030 (but the MJE15030 is easier to find). Also a 2N3440 with a "star" heat sink will work.
TR1 = 2N5401 (PNP 150 volt transistor).
D6 = 1N5228 (3.9 volt 1/2 watt) or 1N4730A (3.9 volt 1 watt) diode.
D5 = 1N4004 diode
D9 = 1N4764 (100 volt, 1 watt) diode. Use 1N4763A (91 volt, 1 watt) diode instead (to increase score display life).
R8 = 39k ohm, 1 or 2 watt flame proof resistor.
R12 = 680 ohm, 1/2 watt resistor.
R11 = 330k ohm, 1/2 watt resistor.
C13 = 0.1 mfd 500 volt (or 250 volt will also work) metal polyester capacitor.

Negative Alpha-Numeric 100 volt section parts to replace:

TR4 = MJE350 or MJE15031 transistor. The MJE350 is a PNP 300 volt transistor, and slightly more robust than the MJE15031 (but the MJE15031 is easier to find). Also a 2N5416 with a "star" heat sink will work.
TR2 = 2N5551 (NPN 150 volt transistor).
D8 = 1N5228 (3.9 volt 1/2 watt) or 1N4730A (3.9 volt 1 watt) diode.
D7 = 1N4004 diode
D10 = 1N4764 (100 volt, 1 watt) diode. Use 1N4763A (91 volt, 1 watt) diode instead (to increase score display life).
R9 = 39k ohm, 1 or 2 watt flame proof resistor.
R13 = 680 ohm, 1/2 watt resistor.
R10 = 330k ohm, 1/2 watt resistor.
C12 = 0.1 mfd 500 volt (or 250 volt will also work) metal polyester capacitor.

HV power supply for an Alpha-Numeric dataeast game.

When installing the new parts, cut the old parts out first. Then use a solder sucker and clean out the circuit board holes. Make sure to install new resistors slightly off the board, to allow for air circulation. Solder all parts on both sides of the board.

When installing the newly fixed board, measure the output voltages BEFORE you plug in the connector CN5 going to the score displays! Output voltage should be between 90 and 105 volts.

Slow Alpha-Numeric Display Strobing Problems.
Sometimes the score displays can have a slow reveal and wipe pattern that leaves only three or four digits on at a time. This can show as three or four digits in the player one and three displays, then the credit display, and finally the player two and four displays. Then the next three or four digits display the same way across all the score displays. You never see the entire display lit. The game will do this in game or game over mode, just repeating the pattern. The score displays do however keep score accurately. Sometimes this problem can be seen in a single score display.

This is usually caused by weak high voltage going to the score displays. Instead of plus or negative 100 vdc, the voltage can be as low as 50 volts. This low voltage can be caused by resistors R8/R9 (39k ohms) on the power supply board, or a bad high voltage capacitor C10/C11 (100 mfd 150 volt), or C12/C13 (0.1 mfd 500 volt metal polyester cap) on the power supply board.

Partial Segment Failures on Alpha-Numeric Score Displays.
Sometimes only parts of a display aren't displaying, and only on certain numbers. For example, the top part of a "0" or "7" are not displaying. It can even be so strange so that the missing segment will work on some numbers or letter, but not on others.

This can be caused by one of the hex input buffers at U6-U8, U1, U5, U6 (14050 or 4050) on the display controller board, which are CMOS chips and static sensitive. These can be checked with a DMM ohm meter. Connect one lead to ground, and the other lead to each input/output pin of the 4050 chips. Any pin that doesn't read the same as the others probably means the chip is bad. Sometimes the chips will need to be tested with an analog ohm meter, and watch for needle "bounce". A pin that doesn't "bounce" like the others probably means that chip is bad.

More Alpha-Numeric Segment Problems: the UDN7180 chip.
If there are still display problems (and the power supply board is working), the next course of action is to check the UDN6118 and UDN7180 chips on the display controller board. The UDN6118's control the strobe pulses, and the UDN7180 control each segment in the display. Usually the UDN7180 is the one that fails (and are unfortunately hard to get and somewhat expensive). There are four UDN7180's and four UDN6118 chips on the display controller board. The UDN6118 and UDN7180 both have the same pinout: pins 1 to 8 are the input side of the chips, and pins 11 to 18 are the output side. Check the schematics to see exactly which chip controls which display when diagnosing these.

Alpha-Numeric Segment(s) Still doesn't work: Check the Resistors before replacing a UDN7180.
Since UDN7180 chips are expensive and hard to find, make sure it really is the problem before replacing any. If the 4050 chip(s) have been checked, and the UDN7180 and UDN6118 chips probed, there is one more thing that should be checked; the resistors coming off the UDN7180 chips. Though it doesn't happen a lot, a bad resistor solder joint or bad resistor itself could be your problem.

Check these resistors with a multi-meter to make sure they are at the correct value. Using the schematics, find the segment letter identifier that are missing. Then follow that segment through the UDN7180 and its corresponding resistor. Make sure that resistor is within 10% of the schematics specified value. If the schematics are not available, check ALL the resistors on the display controller board with a multi-meter. The slightly larger 1/2 watt resistors tend to be the more troublesome resistors.

Alpha-Numeric and Numeric segment letter designations.

CPU board Problems Causing Missing or Locked on Segments.

The resistor (RA) networks are easy to test. Just use a DMM set to ohms (game power off) and put one lead on pin 1 (square board pin) and test all the other pins. Should see 4.7k or 2.5k ohms (all pins on a single RA should test the same value). A pin that is low or high is a problem, but it is not necessarily the RA network that is the problem! If a PIA that feeds the RA is bad, it can make the RA test as bad.

A quick test of the PIAs at 9B and 7B can be done with a DMM set to diode test and the power off. Put the red lead on ground and probe each pin of the PIA. Values should be in the .4 to .6 range for all pins except pins 1,20,22,24,40. If any values are outside that range and the PIA chip is most likely bad. Further tests can be conducted on the PIA using Leon's Test EPROM if desired. Interestingly, PIA 9B can be left out of the CPU board and the game will boot. Of course the display will be largely blank, but this simple test will tell you if a locked-on display segment is caused by a bad 9B PIA. Unfortunately this can not be done with PIA 7B.

Alpha-Numeric segments on a Simpsons (in display test) that are locked on.
The problem with the upper display glass was caused by CPU board PIA 9B.

The same game and CPU board as above (in display test), but now PIA 9B has been
replaced. The "r" segment is still locked on on the lower display glass. Replacing
the PIA 7B fixed the locked on segment on the lower glass.

Will a Williams Alpha/Numeric Display work in a DE Game and Vice Versa?

Rottendog Replacement Power Supplies.
Rottendog Amusements sells a replacement power supply board for Alpha-Numeric Dataeast (and Williams System 11) games, small DMD Dataeast games, and medium DMD Dataeast games. This is a good product at a very low price. The Alpha-Numeric and Small DMD power supplies are essentially the same. I would suggest buying the Small DMD version (version "B" as Rottendog calls it), as it can be easily modified to work in either Alpha-Numeric games or small DMD games. If the "A" (alpha-numeric) version is bought, it can't be easily used in small DMD games. Also the medium "C" DMD Rottendog power supply can not be converted to the Alpha-Numeric "A" or Small DMD "B" power supply.

Rottendog "B" replacement power supply for DataEast and Williams System11 games.

Rottendog "B" replacement power supply section to modify to change between Alpha-Numeric and small DMD games.

To change a Rottendog "B" power supply to be used in an alpha-numeric Dataeast or Williams system11 games, just do the following:

Remove fuse F7.
Change jumpers next to fuse F7: RS10a & RS30b installed, RS10b & RS30a removed.
Cut pin 5 (+12 volts) of connector CN5 on the Rottendog power supply (lower left corner).

On the "C" model (medium sized dot matrix DE games) Rottendog power supply, there was an issue with the 68 volt diodes burning up at power-on. This was due to the game's in-rush current frying the pair of 22 volt 1N4748a (1 watt) Zener diodes at DZ2, and often the LM317 at U1. Rottendog fixed this by changing the pair of 1N4748 diodes at DZ2 to a single 1N4748 diode with the banded side tied to a 10 ohm 1/2 watt resistor.

Rottendog "C" replacement power supply on a Tommy that fried at DZ2 because
of in-rush current. (A kid was constantly turning this game on and off quickly, which
caused the DZ2 diode pair to fry - this also killed the PAL at U16 on the dot
matrix controller board.)

3n. When things don't work: Memory Error (Open Coin Door) and Batteries.

The "open the coin door" message, as seen on 192x64 dot matrix games.

The "open the door" message, as seen on 128x32 dot matrix games.

Replace the Batteries First.
The first course of action is to replace the three AA batteries. This should be done every year, just like replacing the batteries in a smoke alarm. When replacing the batteries, examine the battery holder. If there are any signs of corrosion, replace the battery holder.

When replacing the batteries, in a perfectly working game, make sure the batteries are replaced with the game powered ON. This way the audit and adjustments will not be lost!

Batteries Replaced, Still Memory Error.
From