Deger Design Flipper Circuit.
Starting with Playboy, DataEast changed to a single winding coil (instead of the traditional two windings, one for low resistance high power, and one for high resistance low power hold). The single winding coil had high voltage delivered for the initial power, and lower voltage for the flipper hold. When the flipper button was pressed, 50 volts DC was directed to the single wound flipper coil. When the normally closed EOS switch was opened, the high power was turned off. A low voltage (9 volts) was then supplied to the flipper coil, through a 1N5404 (400 volt, 3 amp) diode, mounted directly on the coil. This allowed the player to hold in the flipper button for as long as desired, without the flipper coil overheating. This circuit was designed by Mr. Kurt Deger, and is hence called the "Deger design".

The Deger Design single winding flipper coil circuit as used on Playboy and Monday Night Football only.

For Playboy and Monday Night Football, the flipper coils had two diodes on them. One was a standard 1N4004 "voltage snubbing" diode. This prevented a backlash of power going backwards through the system, when the flipper coil's magnetic field collapsed. The second diode, a 1N5404 (400 volt, 3 amp), provided half-wave voltage rectification. This second 1N5404 diode was only used on Playboy and Monday Night Football. It was not needed for the next generation flipper system.

Solid State Flipper Board (SSFB) Generations.
DataEast was the first company to use solidstate flippers. This solidstate design again used a single wound coil (instead of the traditional two windings, one for high voltage, and one for low voltage hold). A solidstate version of the Deger design, the single wound coil had different voltages for the initial power and the flipper hold (only one diode, a 1N4004 "voltage snubbing" diode, was used). When the flipper button was pressed, 50 volts DC was directed to the single wound flipper coil. After a short duration (40 milli-seconds, which was not variable), the high power was turned off and 9 volts was left to hold the flipper coil. The lower 9 volts allowed the flipper button to be held, without burning the flipper coil. Note this was different than the solidstate flipper system used by Williams; which used two separate flipper coil windings (a high and low power winding) in their design.

DataEast put out a nice service bulletin on the operations of their solidstate flipper design. This is bulletin number 49, and is available by clicking here, and here.

There were some complaints that the new solidstate DataEast flippers didn't have the same feel as a traditional EOS (End Of Stroke) system flipper. This was because DataEast/Sega's design had a fixed timing (40 milliseconds) for the high voltage (unlike Williams games, whose electronic flippers reacted to the EOS switch, and turned off the high voltage accordingly).

With Jurassic Park, DataEast implemented an EOS solidstate flipper design. Unlike Williams, DataEast's solidstate EOS switch was normally closed. The amount of time the high voltage was turned on to the flipper coil was still fixed, and not controlled by the EOS switch. The EOS switch was implemented for a different reason. When the ball hit an energized flipper bat, and knocked the flipper backwards (opening the EOS switch), the flipper would be pulsed again with the high voltage for the same fixed time (this was accomplished by the EOS switch connected in series with the cabinet flipper switch). This ensured the held flipper would stay in the up position for the player. This was done because of features implemented on Jurassic Park and Last Action Hero. The "Raptor Pit" and the "Ripper" would fire a ball back at the flipper at high speed. This EOS switch was kept for all games after Jurassic Park and Last Action Hero.

A pair of solidstate flipper boards on the left inside of the cabinet, below the playfield. Note the back flipper board still has its factory original cardboard shorting cover. These are flipper board #520-5080-00, as used on WWF and Baywatch.

Solidstate Flipper Board Problems on JP, LAH, TftC.
On the games Jurassic Park, Last Action Hero, and Tales from the Crypt there is a design problem with their flipper boards, #520-5033-03 (3 flippers, used on JP and TftC) and #520-5070-00 (2 flippers, used on LAH). Games with these board use a normally closed EOS switch on the two lower flippers (only). If an EOS switch is broken or mis-adjusted so it is not normally closed, its corresponding lower flipper will not work (but the upper flipper, if the game has one, will work). This problem was corrected with version #520-5076-00 (3 flippers) and #520-5080-00 (2 flippers), as used on games Tommy and later. The newer #520-5076-00 and #520-5080-00 are backwards compatible. Additionally, starting with Tales from the Crypt, the EOS switch became more robust. This newer EOS switch design is easy to identify because of a "bend" near the end of the switch blade.

With the start of Tommy production (after JP, LAH and Tales from the Crypt), the flipper board and EOS switch changed again. The new 3 flipper board (#520-5076-00) is backwards compatible to JP and TftC's flipper board (#520-5033-03). The new 2 flipper board (#520-5080-00) is backwards compatible to LAH's flipper board (#520-5070-00). These new revisions removed the broken EOS switch problem.

The JP, LAH and TftC flipper boards can be modified to work correctly, like the later flipper boards. DataEast service bulletin number 54 describes this procedure. To view this service bulletin, click here, here, and here.

Flipper board evolution:

• #520-5033-00: Robocop to Rocky & Bullwinkle. Supports two flippers, and did not use an EOS switch on the flipper coils (this version was also used as a test on about 100 Playboy 35th games, and 100 Monday Night Football games).
• #520-5033-03: Jurassic Park, Tales from the Crypt. Supports three flippers, and uses an EOS switch on the flipper coils. This board had a design error such that if the EOS switch was broken, then the flipper would not work. Hence it was replaced by #520-5076-00.
• #520-5070-00: Last Action Hero only. Supports two flippers, and uses an EOS switch on the flipper coils. This board had a design error such that if the EOS switch was broken, then the flipper would not work. Hence it was replaced by #520-5080-00.
• #520-5076-00: Jurassic Park to Batman Forever. Supports three flippers and EOS switches on the flipper coils. Can be used on Jurassic Park and Tales from the Crypt to replace #520-5033-03.
• #520-5080-00: Only used on WWF Royal Rumble and Baywatch (plus Apollo13 and Goldeneye, but these are "Whitestar" version games and not covered in this guide). Supports two flippers with EOS switches. Can also be used on Last Action Hero to replace #520-5070-00.

As interesting trivia, the acronym "TY-FFASI" can be seen silk screened on many of these solidstate flipper boards. In the early 1990s, Williams was suing DataEast for a range of things, from board design to playfield plagiarism. DataEast responded with this acronym, which means, "take your fu**ing flippers and stick it".

A special note to the lawyers at Williams gaming, silk screened on many solidstate flipper boards.

Using a Newer Flipper Board on Playboy to Rocky & Bullwinkle.
Games Robocop to Rocky & Bullwinkle (or Playboy and Monday Night Football test games) do not use an EOS switch for the solidstate flipper board #520-5033-00. The newer flipper board #520-5080-00 (which supports an EOS switch) can be adapted to work on these older games. The original board 520-5033-00 is no longer available too, so this modification is important if the entire flipper board on these older games needs replacing.

Here are the modifications steps to use the newer flipper board 520-5080-00 in older non-EOS switch games:

• On the component side of the flipper board 520-5080-00, find the connector labeled CN1. Cut the extruding pin number 10 completely off of the board, so it cannot connect to the wiring harness. For reference, pin 2 is the "key" pin.
• On the solder side of the flipper board, add a jumper wire across capacitor C10. This will essentially short out the capacitor, rendering it useless. This capacitor is located right next to the CN1 pin 10, which was cut in the previous step.
• On the solder side of the flipper board, add a jumper wire between connector CN1 pin 1 and pin 2. For reference, pin 2 is the "key" pin.
• On the solder side of the flipper board, add a jumper wire between connector CN1 pin 6 and pin 7. For reference, pin 2 is the "key" pin.
• On the solder side of the flipper board, there is a trace on connector CN1 that runs from pin 2 to pin 7. Cut this trace with an Exacto knife. For reference, pin 2 is the "key" pin.
• Label the board as modified and for use in games without EOS switches.
The original Sega service bulletin number 103 for this modification (including circuit board pictures) can be seen by clicking here and here.

Power Supply Board (PSB) Generations.
Power supply boards were changed when different score display power requirements were needed (or the number of flippers changed). Because of this, power supply boards are NOT downward or upward compatible! QUESTION FOR JOE: Apparently most power supplies used in Lethal Weapon and later are labeled "520-5047-00 REV X". The "X" in the "REV" seems to be A, B or C. Is "REV C" the same as "520-5047-03"?

• #520-5000-00: used from Laser War to the Simpsons. This power supply was designed to be used with alpha-numeric score displays.
• #520-5047-00: used from Checkpoint to Hook. This power supply was designed to be used with the small (128x16) dot matrix score display.
• #520-5047-01: used on Lethal Weapon 3, Star Wars, and Rocky & Bullwinkle only. These are two flipper, large (128x32) dot matrix score display games.
• #520-5047-02: used from Jurassic Park to Guns N' Roses. This power supply supported three flippers and the larger (128x32) dot matrix score display.
• #520-5047-03: used on Maverick to Batman Forever, and supported the super-size (192x64) dot matrix score display.

Score Display & Display Board Generations.
DataEast had several interesting innovations regarding displays. They were the first pinball company to use a dot matrix display (DMD), and the only company to ever use the super-size 192x64 DMD.

• The first system of score displays was only used on Laser War. It had a master display board (#520-5004-00) which controlled all the displays through other slave display boards. This included two 7 digit alpha-numeric displays (display boards #520-5005-00), two 7 digit numeric displays (display boards #520-5006-00), and one 4 digit numeric display (display board #520-5007-00).
• Four combined 7 digit alpha-numeric displays: used from Secret Service to Playboy. Display board #520-5014-01.
• Two 16 digit alpha-numeric displays: used from Monday Night Football to the Simpsons. Display board #520-5030-00.
• Small 128x16 line dot matrix display: used from Checkpoint to Hook. Display board #520-5042-00.
• Large 128x32 line dot matrix display: used from Lethal Weapon 3 to Guns N Roses. This "normal" sized dot matrix display used display board #520-5052-00.
• Super-sized 192x64 dot matrix display: used from Maverick to Batman Forever. Display board #520-5075-00.

Playfield Power Board (MRB and PPB) Generations.
DataEast/Sega used three different playfield power boards. The PPB board is downward compatible (but not to games with the MRB board!).

• #520-5015-00: MRB board, used from Laser War and Secret Service only. No 50 volts handled by this board (and hence, no TIP36c transistors).
• #520-5021-00: PPB board, used from Torpedo Alley to Jurassic Park. Easily identified because it has six fuses. Handles 50 volt power, and holds five TIP36c transistors which are used to ground the 50 volt coils.
• #520-5021-05: PPB board, used from Jurassic Park (late in the production run) to Batman Forever. Easily identified because it has nine fuses.

SMIG Board.
NOTE: WHAT DOES SMIG STAND FOR? On Laser War and Secret Service, the MRB board did not control ground to the 50 volt coils. Instead, a small board called the SMIG board was used. This board had a relay, which in turn was controlled by a TIP122 on the CPU board. When the relay was energized, this completed the +50 volt ground path to selected coils.

The SMIG board's relay uses a CPU TIP122 driver transistor to turn the 32 volt relay on. This in turn completes the ground path for 50 volts to the desired coil.

Sound Board Generations.
There are several generations of sound boards used in DataEast/Sega games. As the games got more advanced, the sound storage requirements grew accordingly. These sound boards are NOT upward or downward compatible.
• #520-5002-00: only used on Laser War, for part of the production. Used 27256 EPROMs.
• #520-5002-02: used from Laser War to Back to the Future. Only difference between this boards and the prior -00 variant is the size of the sound EPROMs they address. This version used both 27256 and 27512 EPROMs.
• #520-5002-03: used on the Simpsons, Checkpoint, and Teenage Mutant Turtles. This board allowed the two voice EPROMs to be as large as 27010 EPROMs.
• #520-5050-01: used from Batman to Lethal Weapon 3. Allowed the voice EPROMs to be as large as 27020.
• #520-5050-02: used on Star Wars, Rocky & Bullwinkle, and Jurassic Park. This sound board allowed the voice EPROMs to be as large as 27040.
• #520-5050-03: used on Last Action Hero, Tales from the Cyrpt, and Maverick. Very similar to the -02 generation (addressing 27040 EPROMs). No further info known on this board.
• #520-5077-00: used on Tommy, WWF, Guns N Roses, and Frankenstein. This board was a completely new design, and allowed up to four 27040 voice EPROMs.
• #520-5126-02: used on Baywatch and Batman Forever. No further info known on this board.

The Dot Matrix Controller Board.
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. These boards are NOT upward or downward compatible.

• Checkpoint to 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 (Leathal Weapon to Last Action Hero) or #520-5505-01 (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.

Cabinet Changes.
Starting with the game Guns N Roses, DataEast changed the positioning of their legs. Now when all four leg levelers were screwed in all the way, the game should be properely "leveled" at 6.5 degrees (the optimal playing pitch). Any adjustment to the leg levelers should be only for uneven floors. DataEast/Sega did not recommend having the game any steeper than 6.5 degrees.

The backbox design changed with Guns N Roses too. The backbox door swings open from hinges on the right hand side (instead of the left). Also the swinging door is easily removable by just disconnecting the two wiring connectors, and pushing the door straight up off the hinge pivot points.

Starting with Ninja Turtles, the playfield mounting was changed. Prior to Turtles, the playfields had a single pivot point which allowed the playfield to rotate up vertically. With Turtles and all later games, a playfield slide bracket was implemented. This allowed the entire playfield to be pulled forward about ten inches. This allowed for easy repair to the top portion of the playfield (where the pop bumpers reside, for example). After the playfield is pulled forward, it could then be tilted up vertically against the backbox.

---

1f. Getting Started: Game List
Here is the list of games and their system generations. This is important to know before beginning troubleshooting and repair.

Playfield Glass Sizes (and Wide Body Games).
DataEast/Sega made three "wide body" games: Guns N Roses, WWF Royal Rumble, and Batman Forever. Because these game have a larger playfield than all the other games, the wide body playfield glass size is 23.75" by 43" by 3/16" thick. This is incorrectly stated in the GNR and WWF game manuals. The standard size playfield glass, for all the other DataEast/Sega games, is 22" by 43" by 3/16" thick. In all cases, the glass should be "tempered" for safety.

---

1g. Getting Started: The Circuit Boards and How they Work

Overview.
The DataEast pinball system can utilize a total of 16 (8x2) alternating power coils/flashlamps, and 14 constant power coils (of which 6 are "special coils"). All solenoids use TIP122/TIP102 or TIP36c transistors which complete the power circuit to ground.

Alternating Power Coils/Flashlamps (Multiplexing)
The basic concept is a "multiplexing" design. Sixteen power devices are controlled by an alternating two bank relay: the "L" and "R" bank (unfortunately, these banks are also known as "A" and "B"; this is inconsistent in many DataEast manuals). This allows one TIP122 transistor (CPU board TIP122 transistors Q39 to Q46) to control two devices (usually multiplexed between a solenoid and a flasher). Also an additional TIP36c transistor (on the PPB board) can be used in conjunction with a multiplexed TIP122 to drive a higher output device. The eight bank selected transistors can control 16 devices. The bank select relay controls which of the two devices any of the eight driving transistors control. The "L" (or "A") bank consists primarily of coils (except for games Time Machine to Simpsons, see below for details on this). The "R" (or "B") bank consists primarily of flasher lamps.

One thing DataEast/Sega did consistently was use brown wires for the "L" (or "A") side, and orange wires for the "R" (or "B") side (for power, from the PPB board to the device). Also the "L" side is always the same as the "A" side, and the "R" side is always the same as the "B" side.

If the bank select relay is not energized, solenoid power V+ is connected to bank "L". Then only the selected devices can be driven by the driver transistors. There is no power available to bank "R" (usually flashlamps). The "L" bank is usually reserved for coils.

When the bank select relay is energized via CPU transistor Q29, solenoid power V+ is connected to bank "R". Then only the selected devices can be driven by the driver transistors. There is no power available to bank "L" (solenoids). The "R" bank is usually reserved for flashlamps.

DataEast multiplexing of coils and flashers. Power starts at the PPB (right center). If the select relay is not energized, power goes to the "L" (left") side device, the coil. If the select relay is energized, power goes to the "R" (right) side devices, the flashlamps. The other side of each device then returns to the PPB board. For the flashers, this goes through a large sandstone resistor (which decreases the 32 volts down to 12 volts) and a diode. For the coils, this goes through a diode only. The circuit then completes to ground at the CPU board. Here the TIP122 transistor (in this case, Q39) completes the circuit to ground. By the way, "P/O" stands for "Pin/Out".

TIP36c Transistors.
When the PPB board was implemented on DataEast's third game (Torpedo Alley and later), TIP36c transistors were used to control the high voltage (50 volt) coils. These transistors are usually used in conjuction with the multiplexed TIP122 transistors at Q39 to Q46; the TIP122 transistor acts as a "pre-driver" for the TIP36c transistor (there is even a pre-driver for the TIP122 too). This is done to isolate the CPU logic circuit from the high voltage device. So the TIP36c controls the highest voltage, which is pre-driven by a smaller TIP122, which in turn is pre-driven by an even smaller 2N4401 transistor. Every TIP36c must use a TIP122/TIP102 pre-driver.

DataEast multiplexing of 32 volt and 50 volt coils. Here the power starts at the PPB (top right). For the 50 volt up-kicker coil, the power, after going through the coil, goes back to the PPB board's Q5 (TIP36c) transistor and D3 diode. It then goes to the CPU board's multiplexing Q44 (TIP122) transistor, which completes the path to ground.

DataEast usage of a 50 volt VUK (Vertical Up Kicker) coil without multiplexing. Here the PPB board's TIP36c transistor Q2 is predriven by a CPU board's "constant power" TIP122 transistor Q24.

Constant Power Coils.
Transistors Q8 to Q13 (TIP122, special coils) and Q23 to Q30 (TIP122) on the CPU board control constant power devices. These devices are not multiplexed, and hence always have constant power available to them.

The Six Special Coils (Switched Solenoids).
DataEast originally did not have CPU control of the "special coils". DataEast/Sega sometimes called them "switched solenoids" (where Williams called these the "special solenoids"). These six constant power coils comprised the pop bumpers, kicking rubbers (slingshots), and kickback solenoids. They are controlled by TIP122 transistors Q8 to Q13 on the CPU board. The original theory behind this was: since these items needed instant response, having them controlled by the CPU would add enough of a delay to slow the solenoids down. After all, the CPU had to sense a switch closure of a pop bumper, then turn on a transistor which would fire the pop bumper coil. It was believed that this would not happen fast enough for good game play.

The down side to this was when a pop bumper switch was stuck on, the coil would stay locked on (or "machine gun") and eventually either burn the coil or blow a fuse. This special coil system also required a redundant switch. This switch was part of the switch matrix, and only controlled the scoring for this device.

Fuses for the Special Coils.
Each special coil had to have its own fuse to prevent melt-down if the activating switch got stuck on. These fuses were installed starting with Monday Night Football. In games Laser War to Playboy, a fuse holder and a 2.5 amp slo-blo fuse should be installed under the playfield, to the blue wire for each special coil. This was indicated in DataEast service bulletin number 20. It also suggested checking that the pop bumper "spoon" switches were adjusted correctly, the special coils were mounted tightly in their brackets, and the metal pop bumper "yokes" were not broken.

Eventually DataEast changed their mind on these "special coils", and made them CPU controlled. DataEast referred to the CPU control of these coils as "non-reflexive" circuitry. This means, regardless of how long a solenoid's switch was closed, the solenoid would be energized only once by the CPU, and for a pre-determined time. With this design, when a switch got stuck closed, the coil would not lock on or "machine gun" and burn (or blow a fuse). DataEast implemented this "non-reflexive" circuit with Version 3 of their CPU board (Back to the Future).

Reversal of the L & R Relay Sides.
On the games Time Machine to Simpsons (seven games), DataEast designed the power distribution through the L/R relay in "reverse". That is, with the relay unenergized, the "L" side powered the flashlamps instead of the coils. This was a design error! Often the L/R relay would get cold solder joints, and simply not work. This meant if a player started a game, only the flashlamps would light (and the game could not be played). If the coils were on the "L" side of the relay, at least the game could be played (or partially played, but without the flashlamps).

The other problem with this design presents itself when a TIP122 transistor fails (shorts). If any of the related TIP122 transistors short to ground (leaving a permanent ground path for the related device), that "L" side device will lock on (as soon as power is turned on to the game). In the "normal" design, this meant a coil would lock on. This was good; it would be obvious to the operator, and the coil would draw enough power to blow its associated fuse. However with games Time Machine to Simpsons, the "L" side device was as flashlamp. This would lock on the flashlamp, which did not draw enough power to blow its associated fuse. Worse, these flashlamps would get so hot they could melt playfield plastics!

DataEast realized the error they made, and corrected the design so the "L" side connected to coils (as done previously), and not flashlamps.

This page in the DataEast manuals (in this case, Jurassic Park) is probably the most important page in the manual (the diagram here has been abbreviated for space). It shows the coils/flashlamps and their associated drive transistors. This page is in the "Flash Lamp/Coil Test" section, "Game Diagnostics" chapter.

This style coil chart starts appearing around Baywatch. Note the multiplexing coils, and the TIP122 (Q44) and TIP36c (Q5) transistors listed for device 3L. This table was a nice addition to the newer Sega manuals.

Left: A cracked fuse holder on a power supply fuse. This fuse holder must be replaced for reliable operation. Right: New fuse holder clips.

More Reasons to Pull a Fuse from its Holder to Test.
Always remove a fuse from its holder to test it. A fatigued fuse will often fall apart when taken out of its holder. This may not be seen if the fuse is tested in its holder. A fatigued fuse tests 'good' in the fuse holder, yet the the fuse wire can pull away from an end-cap as it heats up. Removing the fuse will also allow testing the tension of the fuse clip, and identify fatigued fuse clips.

Proper Fuse Clip Tension.
Since each fuse has been removed for testing, is a good time to also test the fuse clip's tension. With the fuse removed, use your fingers, and gently push the fuse clip contacts together slightly. If they feel "soft" or "mushy", replace the fuse clip! There should be some tension; if not, the fuse clip is close to fatiguing and cracking, and should be replaced. On solidstate flipper games (Robocop and later), don't forget to check the flipper board fuse clips too. Notice there are two contact points to the circuit board for each fuse clip (top and bottom sides of the board); test continuity from the fuse clips to the board traces on both sides of the board!

A Failed Backbox Lamp Matrix Fuse Makes DMD Games Look Very Broken.
On Checkpoint to Guns N Roses (all games with small 128x16 and mid-sized 128x32 dot matrix displays), there is a unique problem that can make the game look very broken. Random horizontal lines and garbage can be shown on the dot matrix display (but actually the game's CPU has booted correctly and is working OK). On games with the small 128x16 displays, random horizontal lines can appear first, followed by the entire display lighting and staying lit.

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 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.

Interestingly, on 128x32 displays, this problem will only occurs on games with "Babcock" brand dot matrix displays. The 128x32 displays made by Cherry and Dale seem unaffected by a failed 18 volt backbox lamp matrix fuse. This happens because the Babcock displays require 12 volts to function, where Cherry and Dale 128x32 displays do not. This problem can also arise on Cherry brand 128x16 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 were 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 remains unaffected.

---

2b. Before Turning the Game On: Burnt Connectors
Burnt connectors are a major problem on just about all makes and models of electronic pinball games. DataEast/Sega are no exception to this. Burnt connectors can cause small annoying problems (like no general illumination (GI) lights, or big problems like game resets.

Burnt GI Connectors.
Often after getting a new game, after turning on the power, the general illumination (GI) lights don't work and the whole game is somewhat dark. This can be caused because connectors burned.

The GI connectors can get hot and fail. This happens because the molex connectors don't always have enough surface area to handle the GI power requirements. The heat from the connector will cause the solder joints to fatigue which causes resistance (and more heat). The connector pins get so hot they soften the solder. All this causes more resistance, which causes more heat. It doesn't end until the board burns, the fuse heat fatigues and fails, or the connectors pins fall out (or burn!), and open the circuit.

A burnt GI connector on the DataEast power supply. The GI connector got so hot, one male molex pin unsoldered itself from the board! Note the GI relay behind the connector.

The Flow of the GI Power (GI Failure).
DataEast's design for the electrical flow of the general illumination power leads to lots of potential problem areas.

• GI Power comes from the transformer.
• GI Power goes to the power supply board through an input molex plug.
• GI power goes to a relay on the power supply board. This relay can be activated to turn on and off the GI.
• GI power leaves the power supply board through an output molex plug.
• GI Power goes into the PPB board through an input molex plug.
• GI Power goes through a fuse (and potentially cracked fuse holders!).
• GI Power leaves the PPB board through an output molex plug, and goes to the backbox or playfield.

Notice all the molex plugs in the above power flow! Each molex plug is a potential problem in the GI power flow, and can (and does) easily burn. Add to this cold solder joints on the GI relay, and bad fuse holders on the PPB board, and that is a formula for general illumination failure.

Fixing a Burnt Connector.
Fixing a burnt connector requires more than just replacing the connector! The board will also need to be removed and the male pins replaced. Sanding the pins clean and tinning them with solder is a very short term solution. Since the original plating is burned off the pins, they will re-tarnish quickly, and the problem will soon re-occur. If the connector only is replaced, and not the board pins (or minimally cleaned and tinned), the resistance will still be there (from the cold or fatigued solder joints and dirty pins). The new connector will burn in short order.

A crimping tool (top), two different types of pins (bottom left), a new connector housing and male pins. Note the connector pins; the far left two pins are the crimp-on, single wiper type. The two pins to the right are insulation displacement pins, but with two wipers. It's ideal to use the crimp-on style pin, but with two wipers (not shown).

If the game is getting home use, replacing the burnt GI connectors with the white low temp versions is acceptable. Cleaning and tinning the board pins can be done temporarily. These may burn again, after extended use. If the game is being operated (where it's powered-on for hours and hours), replacement connectors should be the black hi-temp molex type (available from Pinball Resource), and make sure new pins are installed.

Crimp-On Pin Connectors vs. Insulation Displacement Connectors (IDC) Plugs.
Insulation displacement connector (IDC) plugs are very convenient for an assembly line or automated procedure to install. No wire stripping is needed, the wire is just pushed onto the "V" in the pin, which cuts (displaces) the insulation to make contact with the wire. But they aren't very good in the long run. Many problems with games are attributed to these IDC plugs. A far better connector uses the crimp-on style of pin. A special tool is needed to crimp them, but the reliability will be much higher. Only use crimp-on pin connectors when replacing burnt ones. Help with crimping connectors can be found at http://www.molex.com/tnotes/crimp.html.

Trifurcon Connector Pins.
Molex makes a crimp-on .156" size female terminal pin called a "trifurcon" pin (not available in the .100" pin size). This style .156" pin differs than the "normal" pin; the metal material is more heat resistant, and it has three wiper contacts instead of just one. The more contact points means the female pin "hugs" the male header pin with greater surface area. I highly recommend these. They are available from Digikey (800-344-4539 or http://www.digikey.com). Specs for these pins can be viewed at http://www.molex.com/product/pcb/6838.html. Compares these to the "normal" connector pin specs at http://www.molex.com/product/pcb/2478.html.

Be aware Molex sells these pins in "strips" or on a "reel". Do NOT buy connector pins this way! Always buy them in "bags" (separated). It's just too difficult to cut them when they are in strips. If a good job cutting them is not done (using a sharp pair of scissors helps!), they won't insert into their plastic housing correctly. Remember to always get the tin plated version, NOT the gold plated pins.

• .156" Trifurcon pins (three wipers): Digikey part# WM2313-ND (Molex part# 08-52-0113, for 18 to 24 guage wire). Price is about $1.02 for 10, $8.50 for 100.
• .156" tin pins (one wiper): Hosfelt (800-524-6464) part# 08-50-0106, for 18 to 24 guage wire). Price is about $0.06 each.
• .100" pins: Digikey part# WM2200-ND (Molex part# 08-05-0114). Price is about $1.05 for 10, or $8.75 for 100. Hosfelt (800-524-6464) also sells these, part# 08-50-0114, for about $0.08 each.

Board Mounted Header Pins.
These are available in several styles. Get the most number of pins available, and cut the header to the size needed. They also come with a "lock" and without a lock. The lock variety is what is used the most.

• .156" header pins with lock (12 pins): Hosfelt (800-524-6464) part# 26-48-1125, about $0.94 each.
• .156" header pins with no lock (12 pins): Hosfelt (800-524-6464) part# 26-48-1121, about $0.70 each.
• .100" header pins with lock (12 pins): Hosfelt (800-524-6464) part# 22-23-2121, about $0.79 each.
• .100" header pins with no lock (12 pins): Hosfelt (800-524-6464) part# 22-03-2121, about $0.72 each.

Connector Housings.
Sometimes the plastic connector housing has to be replaced if it is burnt (in addition to the pins within the housing). Get the connector housing with the most pin positions available, and cut the connector housing down to the size needed.

• .156" white housings (12 pins): Hosfelt (800-524-6464) part# 09-50-3121, about $0.79 each.
• .100" white housings (12 pins): Hosfelt (800-524-6464) part# 22-01-3127, about $0.69 each.

Polarized Pegs.
A polarized peg is a small nylon plug that goes into the connector housing so the housing is "keyed" (keyed plugs can not go into the wrong board header pin connector). It is strongly recommended using these when replacing connector housings.

• .156" polarized peg: Hosfelt (800-524-6464) part# 15-04-0220, about $0.13 each.
• .100" polarized peg: Hosfelt (800-524-6464) part# 15-04-9210, about $0.10 each.

Burnt GI on Time Machine and Before: Service Bullentin.
DataEast's service bulletin number 16 suggests a burnt GI circuit on power supply boards used in games Time Machine and before. They say the GI board traces are not heavy enough to carry the current needed. To prevent a burnt GI section, they suggest the following modifications (these modifications were included on games Playboy and later). Please see the service bulletin by clicking here and here.

---

2c. Before Turning the Game On: Quick and Easy Transistor Testing.
Before ever turning on a new DataEast/Sega game, test all the TIP122/102 solenoid transistors. Do this while in the backbox (examining the fuses and the GI connectors). A blown transistor can really confuse a game. This is the procedure to use, and it only takes about 60 seconds to test all the TIP122/102 transistors:
• Make sure the game is off.
• Put the DMM (digital multi meter) on ohms (buzz tone).
• Put one lead on the ground strap in the backbox.
• Touch the other lead to the metal tab on the TIP122/102 transistors.
• If zero ohms (buzz) is shown, the transistor is bad! (shorted on).

There are a number of TIP122/102 transistors on the DataEast/Sega CPU board:

• Q8-Q13: located at the upper right (special coil transistors).
• Q23-Q30: located at the lower left (constant power transistors).
• Q39-Q46: located at the lower left (multiplexed transistors).
• Q72-Q79: located at the lower right (lamp matrix return rows).

Replace the bad TIP122 transistor(s) with a more robust TIP102 immediately before turning the game on. Replace the associated pre-driver 2N4401 transistor too.

---

2d. Before Turning the Game On: Relays and Playfield Device Driver Board Problems (Cold/Fatiqued Solder Joints).

Relays.
All DataEast/Sega games have a minimum of three relays in the backbox. These relays are a source of common problems on these games. Most common is cold solder joints where the relay is soldered to the circuit board. The relays themselves can also fail however. These relays really aren't repairable; just replace them. Often the contacts on the relay switches will burn together so the relay cannot do its job.

Backbox circuit board relays and their functions:

• PPB/MRB board relay K1: this relay controls the solenoid L/R select circuit, and is controlled by CPU transistor Q29 (TIP122). It is extremely common for this relay to have cold solder joints! If the L/R (or A/B) select coils or flashlamps do not work, this relay should be checked. This relay is a 24 vdc 10 amp relay with two DPDT switches. There are a total of 8 solder points on this relay; two connect to the winding of the relay coil (to turn the relay on and off). The other six contacts (two distinct sets of three contact DPDT switches) are the relay switches. Source number FRL264 P024/02CK, Sega part 190-5002-00.
• Power Supply board relay RY1: this relay controls the GI (General Illumination), and is controlled by CPU transistor Q28 (TIP122). It is very common for this relay to have cold solder joints, or to have relay switch contacts that are melted together. This relay is a 24 vdc 10 amp relay with two DPDT switches. There are a total of 8 solder points on this relay; two connect to the winding of the relay coil (to turn the relay on and off). The other six contacts (two distinct sets of three contact DPDT switches) are the relay switches. Source number FRL264 P024/02CK, Sega part 190-5003-00.
• CPU board relay RY1: this relay controls the flippers (turns the flippers on when a game is started), and is controlled by CPU transistor Q80 (2N4401). The flipper relay is the most reliable of the three, and does not fail often. This relay is a 6 vdc, 5 amp, 4 pole relay with four SPST switches. These switches are wired into 2 pairs (this is done because there can be up to 2 pairs of flippers). There are a total of 10 solder points on this relay; two connect to the winding of the relay coil (to turn the relay on and off). The other eight are for the switches. Sega part number 190-5001-00

Replace Relays with the RIGHT Relay!
All of the replays used in DataEast/Sega games have at least two sets of contacts. If a relay is mistakenly replaced with (for example) a single set of DPDT contacts, it could short the power supply and start it on fire! (I have seen this happen.)

Wrong PPB Relay installed on some TftC Games.
During production of Tales from the Crypt, the wrong PPB L/R select relay at K-1 was installed. This can cause solenoids to activate randomly, and flash lamps to not activate at all. The problem was the PPB's K-1 relay was installed with a 24 volt AC relay, instead of a 24 volt DC relay. The incorrect AC relay had a relay coil winding of about 2 ohms, where the correct DC version was about 650 ohms! The lack of ohms will cause the relay to burn and fail. The relay can be identified easily because it is labeled "24v AC", where the correct relay is labeled "24v DC", or the relay's winding can be measured with a DMM. Games with serial number 101664 or later have been inspected for this problem. If a 24 volt AC relay is discovered, replace it with the correct 24 volt DC relay. This was mentioned in service bulletin number 52.

Playfield Device Driver Boards.
Many DataEast/Sega games use various playfield devices. This includes items such as playfield magnets (used on Guns N Roses for example) or motors. Most of these devices are controlled by an under-the-playfield driver board. The driver board for these playfield devices are very prone to cold solder joints, causing the device to not work, or work intermittently. This happens because the playfield vibrates from the pinball itself, and from the (energizing) coils attached to the playfield. These vibrations can easily cause solder joints on the playfield mounted boards to crack. It does not matter if this board is a relay board, or a newer FET driver transistor board. Both types are very prone to fatiqued (cracked) solder joints.

FET Playfield Driver Boards.
Starting with Guns N Roses, DataEast starting using FET (Field Effect Transistor) driver transistors on some under-the-playfield device driver boards. The advantage to this system was simple; the FET transistors could be "logic driven". This means a 74LS373 TTL (Transistor to Transistor Logic) chip (as commonly used on the CPU board) could directly drive a FET transistor. There is no need for the 74LS373 chip to drive a small 2N4401 transistor, which drove a larger TIP102 transistor, which finally drove an even larger TIP36c transistor (and ultimately the playfield device). The FET transistors meant less components to fail and test, meaning better reliability.

There were some problems with these FET under-the-playfield boards. On Guns N Roses (the first FET application for DataEast), the game used three playfield magnets to alter the trajectory of the steel pinball. The three magnets were activated by an under-the-playfield board with three FET driver transistors. The FETs had large heatsinks, and were mounted upside down (under the playfield). Often the solder joints on the FETs would crack, causing the magnets to fail or lock-on. Resoldering the fatiqued solder joints fixes this problem.

There was another problem with the first generation of FET boards. Instead of the FET driver board being in the backbox close to the CPU board, long wires connecting the CPU data lines to the under-the-playfield FET board were used. This meant any extraneous playfield noise (from the playfield lamps, for example) could cause "interference" on the data lines going to the FET board. On Guns N Roses, this could cause the playfield magnets to lock on, for no apparent reason.

There are two solutions to this problem. First, the under-the-playfield FET driver board can be moved to the backbox, and the wires going from the FET board to the CPU board shortened. Then the wires going from the FET board to the playfield device lengthened. This solves the CPU data line noise problem by making the data wires much shorter. The second solution is to use .1 mfd capacitors between the FET board's pull-up resistors and ground to "de-spike" the interference. This solution does not remove the data line noise, but just supresses it at the under-the-playfield FET board. This solution is easier to implement, but it may not totally cure the problem.

---

2e. Before Turning the Game On: Should the Game be left Powered On?
This is a very common question. After all, arcades leave their games on almost continually with minimal damage (that we know of!). So why not do it with home games?

Although commercial pinball machines can handle being powered on continually, Do not leave homes games turned on when not in use. Here are several reasons:

• Electronic score displays (dot matrix displays and others) on games have a limited life, which is proportional to how much time they have been turned on.
• General illumination circuits will be stressed. Burnt pins and connectors are very common on games that are on for extended periods of time.
• Light bulbs don't last forever, and aren't all that easy to change on a playfield.
• The bulbs, displays, fans, and transformers only attract dirt when on. Leaving the game on means sucking dirt out of the air and depositing it into the machine.
• Heat generated by the general illumination lamps can warp playfield plastics or help delaminate backglass paint.
• Electricity is a precious resource. Conserve it!

Leaving a pinball on all the time can cost much more than any potential damage that could be done turning it off and on as needed.

End of DataEast Repair document Part One.


---

* Go to DE/Sega Repair Part Two
* Go to DE/Sega Repair Part Three