Arcademuseum Seligenstadt repair logs - short version. OK that other repair log got long. Maybe you want it short and concise, "spare me the detail of how you found the problem, just tell me what it was, since I have a machine that has the same symptom(s)". Here we go. For most of these you'll need the schematics since I barely noted the IC number on the board, just the type and/or what it does. Breakout cocktail Symptom: 1) No bricks, 2) player 2 wasn't controllable. Reason for 1) A 7427 triple NOR driving the SRAM's data and R/W lines was bad. Reason for 2) The 4066 switching between the players was bad. New symptom after a while of working normally: Paddles wobble on the verge of unplayable. The 4066 I bridged (only 2 of 4 gates were used) totally died, together with bad edge connector contacts. Symptom: Score of player up not flashing Cause: LS27@M8 I guess Symptom: Picture dim and wobbly Cause: Main filter capacitor (logic board) Breakout upright Symptom: 1) no coin up, 2) wouldn't start a game even if coined up Reason for 1) Board with antenna circuit in machine with crossover coin switch together with probably bad 7404 input inverters. Reason for 2) Bad logic level going into start game flipflop. The IC generating these signals was knackered, replacing it fixed the machine. Symptom: Eats credits Cause: Unknown, probably bad antenna circuit. Works in other machine. Real cause (after the paddles started wobbling to near unplayability): Big electrolytic cap barfed up. Symptom: No sound Cause: Via on the 12V rail flew off Road Champions: Symptom: No picture or solid gray screen. Reason: Address generator (four 7493) first vertical counter IC bad. Symptom: No road, cars or anything besides the score board. Cause: Probably a corroded pin - road came back after jiggling the board together with everything else. Symptom: Sometimes no coin up (but consistent - if it wouldn't coin up, it wouldn't coin up for quite a while and if it coined up, that consistently worked for hours) Cause: Never found. Swapped all concerning chips and problem persists, but it coins up MOST the time. Symptom: Badly misadjusted. No player car. Cause: Badly misadjusted. Used the manual to set player's car spawn position just above the scorebox instead of far below the bottom of the screen. Enemy cars also too far up. Symptom: Bad character in player score. Cause: Bad 7490 in score circuit Symptom: Player car dark gray, used to be white yesterday Cause: Inverted output of the chip that generated the player's car died. Symptom: Road slightly distorted, monitor ripples. Fix: Set transformer to 220V Symptom: Collision detection for offroad died. Cause: Well... the chip that got the signals for Car1, !street and car2-5 was dead. Symptom: Street has no curves, moves as a whole, road markings flicker wildly Cause: Several dead chips: 2 LS123 and the LS73 that generated the !RECI signal for the delay line. Symptom: Monitor dead Cause: TCA 511 monitor controller chip dead. Symptom: Victory music just plays 1 note Cause: 7407 driver IC on sound board dead. Symptom: Finish line not chequered Cause: bad 7486 in finish line circuit Symptom: White screen (bright gray actually). Cars visible, nothing else. Outlines of road can just barely be seen Cause: Bad 9300 (=LS195) in score box circuit Symptom: Competition Cars too far up ahead, not adjustable by potentiometer Cause: Bad contacts in one of the two game relays. Symptom: Looks like !RECI signal is messy but instead of one pixel the road pattern repeats every 16 lines. On the surface it looks like one of the bucket brigades kicked the bucket, but the road moves smoothly over the entire screen and the distortions depend on the driving speed, so Clock or !RECI. Cause: Bad connections all over again... (the machine has been moved) Symptom: No enemy cars Cause: If you guessed bad connections again, you're correct Symptom: Voltages start to ripple again, visibly on the monitor Cause: Maybe the capacitors start going bad... Symptom: Crashing turns on the squeal sound but nothing else crash related happens Cause: Bad 7400 in S7 next to relais (but again it could have been a bad chip socket contact) Super Road Champions Symptom: Blue Screen with GAME OVER INSERT COIN on bottom, nothing else. Cause: On the video board (the one with the color PROM and the Car generator) C1 (LS00) was bad and constantly output CAR2 signal. Symptom: Road wobbling abnormal Cause: 3 out of 4 MM5056N Bucket brigade chips defective Symptom: Cars keep same size over the entire screen Cause: Broken by design. Watched Holiday by Nazareth and their machine looks exactly the same Symptom: Many sounds missing Cause: Two output inverters (CMOS 4069) had bad gates, also the pots are badly corroded Symptom: High Score circuit doesn't work at all Cause1: Bad segments due to bad 74LS48. Can't buy new ones, used CD 4511 instead, pin 5 needs to be tied to ground instead of +5V to work Cause2: High score memory circuit (two LS174) bad Cause3: High score comparator (three 74C85) bad, the C85 have different pinout, there are no 4 bit comparators in existence with that pinout, had to make an adapter. Symptom: Road heavily distorted Cause: One of the replacement bucket brigades died. Symptom: Right player score sometimes jumps forward by large numbers, only sometimes, only randomly Cause: Bad contacts. It has other stuff happening like enemy cars disappearing or player 2 on screen but no player 1, the start man appearing in the middle of a race or the extended music starting with no reason... Solution: Just shake or nudge the machine. Symptom: Entire screen drowns in red (Hantarex MTC90) Cause: Monitor RGB preamp IC TDA 2530 one output stuck high. Ordering a replacement fixed the problem (but the replacement was a mislabeled 2532 and needed pin 7 floating to work) Symptom testing other RGB preamp/sync module: Black screen Cause: 4066 IC defective Symptom still testing other MQ board: Black screen, only faint outlines of graphics visible Cause: 74LS221 on MQ board. Now we have two working MQ boards. Symptom: Pink car stuck on top of screen Cause: Bad capacitors at the NE556 controlling the "drop down" timers Steeplechase Symptom: No ponies Cause: Error - no sync, just white diagonal lines Cause: Error - no more picture Cause: No HSync Cause: The 7474 generating the signals for HBlank has signals that don't go down all the way to logic low so that the chips that receive the signals don't react. Symptom: A horse runs backwards FAST and the machine eats credits almost as fast as you can push the coin switch. Cause: Bad 7411 which controls the horse reverse movement (fall back). Symptom: Can't control the horses Cause: Bad 7414 input inverter. Symptom: One day horses started jumping on their own Cause: Dead 7404 next to the voltage regulator Symptom: Doesn't get faster Cause: 7404 in acceleration circuit had one gate bad. Symptom: No horses except #7, which moves over the entire screen and does not stop when it reaches the finish line, instead wraps around after some seconds. Cause: Credit button jammed stuck. Symptom: Horse 3 finishes the course in a fraction of a second without the game being started Cause: 7404@K3 (signal name FWD3) PONG (Mirco Games Challenge 4 Player) Symptom: Doesn't coin up the second time. Only 1 and 2 player modes works, no 4 player mode. Cause: The 7439 receiving the 1-coin-up pulse together with the coin-up pulse was bad. Symptom: Picture starts wobbling when hot Cause: Bad diode in rectifier Symptom: Burns fuse when hot Cause: The big electrolytic for the 5V must've been the cause. Cause can be narrowed down to the chassis. Replacing the regulator board didn't fix the problem. Symptom: Again, only 2 can play Cause: My minimal invasive credit button hack stuck closed. Why it is possible to reliably coin up the game in Game Over mode but not be able to coin it a second time when the switch is stuck closed the entire time is beyond me. But once I got the switch aligned right, everything worked. Monaco GP Symptom: Score board malfunctions, overwrites all scores with the last one Cause: A single socketed IC was rammed in the socket so that the pins bent. (don't know which one, so this is pretty useless information...) Also the board has lots of dry joints. Symptom: Sometimes extra score entries with garbage in the tens digit are generated Cause: Bad RAM for the tens digit. Symptom: But instead of sometimes, it should have happened always Cause: LS00@13H had an output stuck high Symptom: No ice (slip zone) sound Cause: IC54 (LM324) bad Symptom: Only one Other Car sound and it's slightly off Cause: Another LM324 modulating the secondary frequencies onto the sawtooths bad (IC11 or 10) -Next machine- Symptom: Crappy picture Cause: RGB amp potentiometers turned far into overdrive (distortion) Symptom: Doesn't coin, steering doesn't work Cause: Turns out, all coin switches whether needed or not must be present Symptom: Time stuck at 90 Cause: Bad tantalum in time circuit right next to the pot and the 555 Symptom: Lo-Fi noise (sounds like a repeating sample, motor or explosion) Cause: LM324 generating the envelope for the explosion bad (tricky because it worked, it just didn't turn the explosion off completely) Symptom: Own car engine noise never changes frequency Cause: Bad tantalum in motor frequency circuitry Symptom: Gear shift glitchy, kinda weird, you can go from 0 to 100 in a few frames if you jiggle it slightly Cause: Unknown, but an LS193 had an output stuck high and an LS157 was totally bad but since I was hunting a different problem, I didn't look for that (also weird to reproduce) Symptom: No ice (slip zone) Cause: Again, a bad LM324 (found by Oscilloscope+Piggybacking) Symptom: Some cars move right only until they hit the border of the screen and wrap around to the left Cause: IC89 (LS157) had an output stuck high -Next machine- Symptom: Some cars move right correctly but they jerk left 16 pixels at a time Cause: Several LS161 counters for the enemy cars totally failed the In-Circuit Chip tester, this time no MID LEVEL but bad outputs. It was five in total. They tested OK out of circuit, so I put them back and that solved the problem. Symptom: Time stuck at 90 Cause: Bad 555 for a change. The tantalum was OK. Symptom: Monitor measures short circuit Cause: Deep discharged mains filter cap. Symptom: Scoreboard completely dead Cause: Blown fuse (no cause found) Symptom: Fanfare way out of tune Cause: Someone turned ALL the tuning potentiometers (6) to maximum. Symptom: Ice (slip zone) sounds is a high-pitched whine that hurts Cause: Potentiometer for Slip zone was set incorrectly Symptom: No ambulance sound Cause: Signal was lost in a bad LM324 Symptom: 1st place shows up in 5th place, 2nd place is completely dark, 1st place display is very bright and appears to be showing two scores at once Cause: Bad LS390 controlling the LS259 which decides which line of score to enable Cause2: Bad LS16 driving the line driver transistors Symptom: After a score is made, one digit of the other scores is garbage Cause: Bad LS174 (there's 20 of them, one for each digit of the high score table) ---some assorted boards--- Symptom: Clock/Scenery board produces only colorful moving horizontal lines Cause: Bad 74161 somewhere in the middle Symptom: Car board repeats cars horizontally (Y-direction) Cause: Two bad 74157, one is the one in the upper left corner if the edge connector is on the right. The other one was in the same line as the first ROM. Symptom: Car boards produces unsynced mess of pixels Cause: Many chips bad. I pulled two bad 157 and a 161 and no difference. TANK: Symptom: 12V rectifier diodes charred Cause: Shorted capacitor Symptom: 5V doesn't regulate (at ~3V), still no 12V. Cause: No 12V. Traces for 12V have both burned through. Symptom: Picture lacks almost everything. Scores are solid squares, minefield is arranged in a clean matrix, two pixels slowly flicker around the screen. Cause: Main ROM doesn't work: Reason: No -12V. Cause: 741 op amp bad. Symptom: No right channel. Cause: LM380 amp bad. Symptom: Player 2 doesn't stop flashing, Player 1 can't shoot Cause: Bad 556 timer. Symptom: Killed the machine. Backgrounds repeat and look weird. Cause: Killed two of the 74153 controlling the big ROM. Symptom: Player 1 can't aim at certain angles. Cause: 7474 on E2 (left board) controlling the 9316(=74LS161) which is responsible for the bullet speed in one quadrant was dead. Symptom: Player 1 can't shoot left (shots get mirrored) Cause: Another 7474 but I forgot where. One right and I think three up from E2 ---Next machine--- Symptom: Completely dead Cause: Transformer center tap edge connector burnt away completely Symptom: +5V not regulated, rippling at around 3V. Cause: The final regulator transistors were both bad but tested OK out of circuit. They had gain, but couldn't drive any current. Symptom: Equivalent of garbage screen. Backgrounds keep repeating in both directions, the tanks move right super fast. Cause: Every single of the five LS153 addressing the ROM was bad. Also an LS86 had one bad output. Quadrapong: Symptom: No sync, picture looks weird. Playfield is quadrupled in H-direction and doubled in V-direction. Monitor appears to be working fine. Cause: 7493 in the address generator was bad. Missile-X: Symptom: No missile Cause: A bad 74161 counter Symptom: Tank in upper line goes too fast Cause: Bad 7474 bottom left one up from the big ICs which I thought were ROMs Symptom: Only one tank in upper line Q*Bert: Symptom: No Knocker Cause: Knocker coil bad Symptom: Horribly unstable Cause: Power supply board didn't have dry joints, it WAS a dry joint. Cause 2: 5V fuse holder was horribly corroded. Symptom: No sound Cause: 30V fuse blew (reason: when the game crashed because of the corroded fuse holder and the dry joints in the power supply board, the knocker coil turned on and didn't turn off and some idiot put in a 7 amps fuse for the knocker fuse and the sound board fuse was 2 amps) Symptom: Red randomly turns off. Cause: Dry joints on monitor PCB. Symptom: Plays the wrong sound effects for some numbers of sound effects, most sounds are correct though. Music gets stuck on the first note. Cause: Sound board 6502 CPU was knackered but worked well enough to not be suspicious. Symptom: No vertical deflection Cause: A dry joint slipped my attention when resoldering the monitor. Donkey Kong Junior Symptom: No VSync Cause: 74LS123 in the vertical sync circuit (next to the V Position pot on the video board) dead. Symptom: Audio distorted, blaring at maximum volume, volume pot barely does anything Cause: Audio Amp died. It runs on 100V+120V. Maybe I test the transistors in a transistor tester someday. Another DK Jr Symptom: No sprites except eyes of the crocodiles Cause: 5V was at 4.5V. At 4.9V, all the sprites were there but some were glitchy, which is normal for DK HW. Symptom: No digital sound Cause: Ripped resistor Some more DK Jrs Symptom: No backgrounds, only sprites Cause: The sprite EPROMs read almost blank. Stored in a sunny place... Symptom: Some Klaptraps color the screen red to their right Cause: Scratched trace Symptom: No digital sound (but if you hold a finger to the DAC and the other to the volume pot, the digital sound can be heard) Cause: Bad DAC08 Symptom: No digital sound (no such luck here) Cause: Missing sound ROM Regular Donkey Kong Symptom: No walk sound Cause: Bad NE556 at 8N Another regular Donkey Kong Symptom: Picture jitters horizontally (monitorwise) when many sprites are onscreen, the monitor is not the cause Cause: Apparently, Donkey Kong has some timing issues. There are optional deglitching caps all over the address generator. The jitter is so subtle that I can't even find someone to describe the picture while I'm changing caps and trying things on the back of the machine. Since I know what to look at, I found that Donkey Kong Jr. has the same problem, but at least one of our own DKs is rock solid. Its monitor isn't, so we might swap out the boards. Another Donkey Kong (Hi-Score Hanover) Symptom: Bad characters in text strings, weird playfields, plays the weird playfields Cause: CHR ROM at 3N had an internal short on address line A7. Radarscope Donkey Kong Symptom: Garbled screen, rarely works, if it does, the screen is still garbled Cause: Bad IC sockets Symptom: No sound Cause: Test wiring harness missing the power cable for the sound board Symptom: still more bad IC sockets, I got the garbage screen again, but sound works. Crazy Kong bootlegs (lots of them) Symptom: Garbage screen/crashes/flaky Cause: They all have bad ROM sockets. One had bits flipped due to light hitting the EPROM Xevious Symptom: Blue screen Cause: Bad rectifier Symptom: Blue screen again (after 5 years) Cause: Bad 2N3055 in AR3. Cause for the color: Bad pots in monitor. N-Sub Symptom: Solid amber screen. Symptom: Only 3.7V on the 5V rail Symptom: 0V on the 12V rail Cause: Shorted tantalum cap on the 12V rail Action: Swapped beefy power transistor Reaction: Now 4.5V on the 5V rail Game now runs. Symptom: Has a solid continuous tinnitus beep on the speaker Cause: (found out later that the 5V had bad ringing on them). Makeshift cure: Put a cap on the +-15V amplifier supply. Symptom: Crashes frequently, still only 4.5V on the 5V rail Cause: Ringing on the 5V rail Cause: Bad deblocking cap on the input of the 7805. Symptom: Crashes after a few minutes Cause: One of the small customs lost its heatsink Symptom: Crashes after moving machine Symptom: No red Cause: Corroded IC sockets Sega Vic Dual (Head On+Space Combat) Symptom: Nothing, no picture Cause: Probably bad sockets, but also an LS253 with an input shorted to GND and an LS374 with a bad output Symptom: Garbage screen Cause: Bad RAM sockets Night Driver Symptom: No steering Cause: Bad opto amplifier transistor Symptom: Only 2 of 3 tracks can be selected Cause: Bad LS151 (or was it 153? Anyway it was connected directly to the knob) Symptom: Gear shift works intermittently, certain gears don't work sometimes Cause: The wires just keep breaking off... Symptom: Black screen with angry noises (screech/crash) Cause: Burnt off power connector. Afterburner, Enduro Racer, 720 degrees, Super Hang on Symptom: Variable amounts of graphic glitches and/or frequent crashing Cause: Bad contacts, wiring power losses, dry joints (Monitor). Symptom: White screen with retraces Cause: (Hantarex 900E) White concrete resistor in power supply ran out of solder. Enduro Racer: Symptom: Steers to the right, not controllable, weird values in self test Cause: Bad ADC Symptom: Randomly crashes Cause: Bad interconnect ribbon cables. Missile command Symptom: Colorful retraces Cause: VBlank too short. A 7404 which creates !VBlank was bad. Symptom: solid red screen, machine dead Cause: Unknown, probably bad contacts. Problem went away and never came back. Another Missile Command Symptom: Garbage screen with flickering pixels Cause: Bad CPU socket Symptom: Self test shows RAM 4K being bad. Only 4K mind you Symptom: After swapping 4K, self test now shows 4J being bad Cause: Buggy self test flags every RAM good after the first bad one. Bubble Bobble Symptom: Doesn't coin up Cause: Security Chip. Using one from a bootleg fixed it but of course the RNG doesn't work now. Symptom: Weird colors that are different every time when switched on Symptom: Crashes after game start (immediately) Cause: One address line from the CPU to a PAL had a broken trace Symptom: Plays now, but no sound Cause: Sound CPU also had a bad trace, a fix was in place, but only worked when measured offline. As soon as you stopped measuring it, it stopped working. Hard to find. Another Bubble Bobble Symptom: Boots to black screen, CSYNC changes between wonky and "this is not a sync signal at all" Cause: Bad LS08@IC27 Symptom: Visible lack of 16H on character layers (sprites are ok) Cause: bad VRAM Symptom: Now every right half of every character is missing every second scan line Cause: bad VRAM Symptom: Almost all the signals are gone and the VRAMs toast themselves Cause: Now the VRAM is definitely bad. Also the 74S74 in the corner died. Symptom: Crashes when trying to start the game (going to Attract Mode) Cause: Bad work RAM (very subtle fault) Symptom: Some colors wrong Cause: Only one of the VRAMs survived the toast. Symptom: Weird enemy behaviour in the higher levels Cause: Maybe that one VRAM? But that looks good. Won't fix. Scramble Formation (bootleg) This is Bubble Bobble hardware. Symptom: Good ol' Garbage screen. Both main CPUs appear to be dead, BUS is idle even though HALT and RESET are high. One CPU has WRITE stuck LOW!? Cause: Both CPUs really were dead. Nothing else. These are GoldStar, I think. The Sound CPU worked though. Symptom: Static "dead CPU" garbage screen Cause: LS157 in chip select logic had an output stuck low Symptom: One of the graphics layer consists of garbage - all sprites glitch Cause: One SRAM near the ribbon cable Symptom: Sound effects/music degrade. Won't stop playing the music, won't start playing the music, plays wrong sound effects or doesn't play them at all when getting warm Cause: (sadly the problem hasn't turned permanent yet so hard to debug, but it is pretty certain that the communication between the main CPU and the sound CPU is the problem as when the music is running, it'll be playing forever without losing anything) Symptom: No music/sound effects Cause1: 12V for amplifier has a bad contact and only works sometimes Cause2: Maybe bad LS74 (failed in circuit but passed out, the replacement also passed in circuit) (NOTE the problem with wrong sound effects persists) Symptom: Stopped scrolling, screen is scrolled about half way but solid (you can see the graphics being reloaded), sprites corrupted, you can get the background layer to scroll by half of the screen by tweaking the +5V Cause: Two LS169s for the most significant scrolling bits are bad (even though they test OK out of circuit, but replacing them fixed the machine) Real cause: (Problem came back after a few weeks) The only LS175 was bad Scramble formation Monitor Symptom: Arcing inside the neck Cause: Tube very likely went to air Symptom: Burns through traces with the replacement tube Cause: Tube has internal short between heater and ground pin. Reason (found much later): Replacement tube's socket doesn't have a connection on that pin at all Fix: Redesigned the heater secondary which also generates the -20V to work with a short in that place. Symptom: Very unstable picture brightnesswise. Sometimes no neck glow. Cause: Completely f**ked up tube socket Symptom: White screen with no picture at all Cause: Replacement tube socket has no connection to the ground pin, so no short to ground, causing the heater secondary to float, causing the -20V to vanish Fix: Created another short circuit... Symptom (monitor works fine after 3/4 years of trying to fix it, then dies after just a few hours): No vertical deflection Cause: TDA3651 vertical deflection IC Symptom: Super shaky wobbly picture Cause: Main filter capacitor bad Symptom: After working for a few seconds, FI fuse blows Cause: Replacement cap shorted against something metal Symptom: Fries voltage regulator transistor and keeps shorting it Cause: A resistor got stuck below a heatsink, rubbing off paint and shorting to it. Symptom: Vertical deflection hyper unstable, vertical lock is somewhere different every time you turn on the machine Cause: I guess the TDA2579 deflection IC Symptom: No green, barely any blue Cause: Green driver transistor on neckboard dead, blue one out of spec NOTE: With a 60-in-1 board, there is no jittering of the picture at any temperature. Picture is rock solid now. Space Invaders cocktail Symptom: Frequent crashes when all the machines in the arcade are running Cause: Reset pin on CPU made no contact because of bad socket. Symptom: Still crashes frequently, but at least now gets resetted by the watchdog. Cause: Bad voltage regulator PCB (exact cause unknown), replacing it with a switcher fixed it. Symptom: Some sounds don't work Cause: Bad LM3900s Symptom: Likes to crash, sometimes Invaders produce white lines Cause: The chip sockets are bad Symptom: After CPU socket swap no base explosion sound (probably unrelated) Cause: After like 2 years another LM3900 died. Symptom: Sometimes stripes through the invaders Cause: Bad interconnect to daughterboard. Space Invaders upright Symptom: Invader graphics ingame have a solid white line on top and bottom, they go right, drop down one, go left, drop down the entire screen and it's Game Over. Cause: Bad LS151 in the barrel shifter on the daughterboard. Symptom: Most sounds don't work Cause: More bad LM3900s. Space Invaders upright (Hi-Score Hanover) Symptom: Crashes frequently with graphical garbage. Cause: Bad ROM sockets, bad power connection Symptom: Still crashes, but less often, maybe once every 5 hours. Cause: Unknown. Bad ROMs maybe? Symptom: No Shot sound and no UFO hit sound Cause: Bad LM3900s at M4, M5 and L4. Space Attack (SI bootlegs by Lich) Symptom: Dead, dead Cause: Both bad 8080 CPUs Symptom: Dead, dead, no BUS activity, swapping the ROMs brings the BUS back but still dead Cause: Unknown, maybe external interrupt required? Symptom: Crashes, erratic white pixels every 16 positions Cause: Bad RAM. Symptom: Crashes (later), rare erratic black pixels someplace. Cause: More bad RAM (very subtly bad) Space Invaders (Geiger Electronics bootlegs) Symptom: Dead, dead Cause (Machine 1): Bad ROM#0 Cause (Machine 2): Bad ROM#0 AND bad multiple bad RAM Galaxian Symptom: No picture Cause: LS161 in video address generator dead (first horizontal) Symptom: Picture is double, missing every 4 columns, game runs at double speed Cause: Another LS161 in video address generator dead (first vertical) Symptom: Many colors wrong or missing Cause: Video signal lower than MAME machine, had to turn up TVs brightness Cause: 2 Color PROM address lines stuck low Cause: Bad 74LS174 next to the color RAM (bottom left on board) Symptom: Hum and only 3V on 5V rail Cause: Beefy power transistor died. (machine now works) Symptom: Dead with colorful garbage on screen that doesn't change. Cause: Bad LS139 near CPU Symptom: Colorful garbage on screen, doesn't change but turns starfield on and off, flips picture and randomly pulses coin counter Cause: Damaged A13 trace to LS139, also bad ROM sockets Symptom: Flickering monochrome garbage on screen, stuck in reboot loop. BAD RAM 2. Cause: BAD RAM 2 was correct. The 2114 tile RAM WAS defective, but barely visible. The RAMs work fine in an Emerson Arcadia 2001. Symptom (different board, no daughterboard, just 2 ROMs): Entire screen filled with white 0 Cause: Both ROMs bad. UniWar S (Galaxian HW) Symptom: Garbage screen, watchdogging Cause: Bad Z80 ---next machine--- Symptom: Garbage screen Cause: Long legs on blocking cap on daughterboard shorted A7 to GND Symptom: BAD RAM 3 Cause: The 2101s were both bad. Ladybug (Galaxian HW) Symptom: Garbage screen Cause: Bad Z80? Symptom: Flickering garbage screen Cause: Bad work RAM (corroded+lost legs on removal) Moon Cresta Symptom: Very dead Cause: 6MHz missing Cause: No output from LS368 Cause: Bad output from LS107 next to it Cause: Several ICs were shorting the 6MHz clock to ground. Symptom: No watchdogging Cause: Bootleggers had WD DISABLE hardwired Symptom: Garbage screen, sometimes self test (but screen is just white) Cause (for sometimes test): Four of the program ROMs were missing! Cause (for white screen): More ICs had Inputs shorted to ground, disabling other ICs. Also all four LS194 were bad. Also two of the LS157 below were bad as well as the LS86 next to the line RAM. Symptom: Still video weirdness Cause: Bad LS08, LS174 in line RAM circuit. Symptom: Garbage screen with different solid color backgrounds Cause (for the solid color backgrounds): Half the CHR ROMs were bad Cause (for Garbage screen): LS365 CPU buffer cracked trace from enable to GND (btw. the board works now) ---Another Moon Cresta--- Symptom: Only brief flash of video, repeats, no watchdogging Cause: One of the CHR RAMs bad Knock Out (on original NAMCO Galaxian) Symptom: 60 different garbage screeens per second Cause: Bad connections to ROM daughterboard. Crazy Kong (on Galaxian HW) Symptom: Garbage screen. Cause: Bad Work RAM Symptom: Garbage screen with signs of life, trying to boot, some code runs (shows Mario and "P1" and "L=00") Cause: Bad Character RAM Symptom: Same as before, but on a blank screen instead of garbage Cause: Bent leg on one of the frighteningly tall hack towers of stacked chips Symptom: Mario wraps around when reaching the middle of the screen Cause: Bad 2101 sprite RAM. Symptom: Every 32 lines there is a bad line of colors Cause: One of the Line RAMs is missing Pac Man (on Galaxian HW) Symptom: Garbage screen Cause: One work RAM and one Char RAM bad Symptom: Lines of bad/glitchy color Cause: One line RAM has bad/dirty outputs, no clean LOW. Symptom: Colors weird Cause: It's a hack of a hack and missing some color resistors, has others added on top Galaga, Pole Position Symptom: Dead/crashes constantly/watchdogging Cause: Corroded legs on the Namco custom chips. Cause: 4.2V on the 5V rail due to bad contacts and misadjusted PSU Pole Position: Symptom: Garbage Screen Cause: Battery acid killed one trace and the PAL below the Sound CPU. Symptom: Stopped working after half an hour Cause: No +5V on any board Cause: Bad rectifier in Power Supply, also all the wires are horribly corroded one trace had to be bypassed Symptom: After a few more hours, only a static garbage screen, albeit with a BLACK background (Pole Position usually shows a garbage screen with a blue background). Symptom: No watchdog Cause: Two more traces eaten away. Fixing the watchdog (by experimenting with resistors until the curve on the scope looked somewhat like it should) set the machine to blue garbage screen. LS138@8D next to the Z80 was the bad guy this time. Symptom: Some notes of the melody sound weird/are out of tune Symptom: Explosions and some billboards have pixelitis Cause: Not gamebreaking, so low priority Symptom: Picture too "hot", could be fixed by turning down the contrast, only this monitor doesn't have a contrast pot! Solution: Monitor doesn't terminate video signals, so 470 Ohm to ground per color channel did the job. Symptom: Keeps dying Cause: Power harness wiring corroded Pole Position 2: Symptom: Good old garbage screen Cause: the bigger of the two Z80 ROMs had the Ground pin bent out Symptom: Screen full of white "W"s, RAM 6 ERROR Cause: Probably a bad contact. When I put the board back in, it just worked Symptom: Dead again with garbage screen, plays weird sounds, does exactly the same every time you turn it on Cause: The CPU just died. Symptom (replacing known working video board with original one): RAM 0 Cause: It really was RAM 0 Symptom: RAM 3 Cause: Jep. Symptom: Still RAM 3 Cause: Machine is picky with exact type of replacement Symptom: RAM 8 (intermittent) Cause: Bad socket. Symptom: RAM 31 Cause: RAM 31 lost a bit Symptom: RAM 46 Cause: well... I must say this machine has a great self test Symptom: Garbled sprites Cause: LS283 at 9M/10M (or thereabouts) bad Symptom: Every other sprite scanline missing Cause: RAM at 9F/10F which the self test can't get to Symptom: Sometimes glitchy sprites Cause: Bad socket for one of the customs Symptom: Colors are weird (they weren't an hour ago) Cause: Color PROM must've something weird happen to it internally (all inputs react, all outputs toggle, but replacing the PROM fixed it) Pole Position 2 (Namco? Bootleg?) Symptom: Bluescreen Cause: One of the Z8002s was bad Symptom: Self test passes in an endless loop, never goes into the game Another Pole Position 2 Symptom: Instantly draws the "Garbage Road" screen Cause: Bad 7C PAL, only the CMOS RAM /CS output was bad. Galaga bootleg Symptom: Watchdogs at initial garbage screen Cause: Empty socket where the custom replacement daughterboard should be Symptom: No coin up (the rest of the board appears to work fine) ---next machine--- Symptom: No sprites Cause: Sprite color PROM was barely hanging in a very damaged socket (the chip survived tho) ---next machine--- Symptoms: Stripes through sprites. only every other scan line has sprite data. Cause: Bad line RAM (2148 the one before the LS298) Stargate Symptom: Dead. Worked-fine-yesterday. Cause: Burnt-off contact pins Symptom: No sound Cause: Sound board rectifier bad. Symptom: Random restarts Cause: Unknown, I accidentally zapped the machine next to the CPU, but the logic signals all look pretty healthy Symptom: Dead again Cause: 5V rectifier shorted. Symptom: Says error 35 Cause: (they do start counting at 1, not 0): Row 3 column 5 RAM chip defective Symptom: Stuck in rug pattern with glitchy graphics Cause: Missing 12V due to bad LM723 Symptom: 1-2-1 error with flickering garbage lines on screen Cause: Bad 7410 at 4 what was it R? There's only two 7410s there, the outer one was still good. Joust Symptom: Starts 100 pixel of rug pattern test, then crashes Cause: Same as Stargate. Measured 4.2V on the logic board. Should be near 5. Of course the contact pins were corroded. Symptom: Inputs randomly fail. The machine is no fun with two players if one player can't flap half the time and the other can't go left. Cause: Pins corroded so badly, it was impossible to get a reliable connection. Fix: Soldered the inputs directly to the board. Sorry, but now it's playable. Defender Symptom: 4 LEDs flash twice but no boot. Rug pattern turns to vertical lines on scan line 200 (or thereabouts), machine redoes rug pattern over and over again Cause: Decoder ROM #3 responsible for Rug pattern fault Cause: Missing ROM in ROM board. Symptom: Machine not doing anything sensible anymore, completely nuts, wildly flashing LEDs, sound board trying to tune into a couple AM radio stations at once, monitor displays dead (pre-rug-pattern) garbage with CGA snow that flashes in time with the watchdog Cause: Unknown. Almost all 7400 logic passed, those that failed passed out of circuit, two chips are marginal. Knowing that 1980's Signetics are problematic in Defender, they can go bad and still the outputs look OK, I will swap some of these. NOTE: Changing the single wipe chip sockets didn't do a thing (but will improve reliability a lot once I fix this thing) Cause: Joystick PIA board causes the machine to crash before even getting to the rug pattern. Symptom: Some pixels in the rug pattern flicker and the machine says RAM ERROR Cause: Had to go to the service menu to get the answer: Code 36, so Row 3 Col 6. Symptom: Rug pattern, then BAD RAM with too many bad pixels for one RAM to be bad, then reboots Cause: -5V rail missing Cause: 7905 regulator bad Another Defender Symptom: "RAM FAILURE", black jailbars Cause: 7432 in RAM circuitry Super shot Symptom: Dead Cause: 3.8V on the 5V rail Cause: Bad 7805 regulator Symptom: Starts with 1 credit, counts credits but doesn't coin up and won't react to start button Cause: Sound board pulls coin signal low Cause: Unknown. Cut the trace and wired the coin signal directly to the coin button. Symptom: Hard to hit stuff Cause: Cun badly calibrated. The two pots behind the coin door control offset, the two pots on the mainboard control amplitude. With a colleague I was able to calibrate the gun good enough. Symptom: No sound Cause: Dead LS374 in the sound bus driver circuit caused by a shorted 7474 on the sound board as well as a low impedance input of a shot 7410 on the sound board. Symptom: A few sounds missing Cause: Lacking the schematics I swapped both sound board ROMs and didn't see they were completely different. Also another LS374 was dead. Video Pinball (waterlogged...) Symptom: No picture Cause: Bad LS157 next to CPU Cause: Bad LS161 in video address generator (the third one). Symptom: Plunger doesn't work Cause: Wrong resistor on plunger board. Replaced with 10k. Symptom: Picture now is an unsynched mess Cause: Another LS161 got burning hot Symptom: No bell Cause: LS93 in bell circuit dead Symptom: No ingame sound Cause: LS161 in sound circuit dead Symptom: Ingame Sounds wrong Cause: other LS161 in sound circuit board not working properly Symptom: Playing with itself, then constantly rebooting, only test seems to work ok Cause: Input pin to unused Chip Type Selector got too sensitive. Symptom: Knackered chip now fails completely, machine dead with garbage screen Changing the knackered LS244 chip in the input circuit didn't do anything, but the chip verified bad. Real cause: LS42 next to CPU died. Just measuring the signals brought the machine back to life when it was working intermittently, now this chip has only garbage on its outputs. Symptom: Plunger not working, only one of its signals arrives at the board, jiggling the connectors didn't help Cause: Taking the plunger unit apart brought it back - huh!? NOTE: The plunger fails from time to time... Symptom: Unstable, keeps crashing, draws play field elements in random positions Cause: Probably bad power connection. When I jiggled the edge connector, it came back rock solid. But it keeps happening. Maybe one of the rectifiers or the mains regulator is on its way out. Symptom: Plunger died Cause: The LS14 I swapped years ago had an input die Symptom: Plunger still very flaky Cause: Maybe a weak transistor Symptom: Plunger still not working all the time Cause: Bad edge connector With some trial&error I put a 620k resistor from the base to (I forgot GND or +5V? I think it was GND). Now the transistor switches harder. Symptom: Picture starts tearing, then loses sync completely Cause: Bad 161 at was it M2 or M[max-1]? Video Pinball Monitor (Wells Gardner 19V1001) Symptom: Sometimes no vertical deflection Cause: MPU131 special transistor bad Symptom: Sometimes no horizontal oscillation Cause: Bad transistor in sync separator, bad transistor in H-oscillator and no idea what else, because it still sometimes doesn't work. Symptom: Screen wobbles when all the other machines are turned on and sometimes just by itself, replacing the mains cap didn't do a thing. Half a Cause: There was a broken wire installed from the factory going to the mains filter cap, giving intermittent connection. But replacing the mains filter cap as well as that bad wire only made it worse (now I can shake the machine as much as I want and the picture keeps wobbling) Cause: Bad diode in voltage regulation circuit. GDR PolyPlay Symptom: No picture Cause: The middle board (Video board with 2k video RAM and 2k video ROM) dead Cause: Bad russian LS193 equivalent Sympom: Screen is a glitchfest Cause: two of the four 2114 video RAMs were bad Symptom: Some custom characters are missing red at their rightmost pixel Cause: On the character RAM/IO Board one of the 48 character RAMs was bad. Symptom: Picture squeezes downwards until it takes up a few cm on the bottom of the screen when warm Cause: TDA2030 knockoff in the vertical deflection dying Symptom: Picture starts tearing up a few seconds after turning on, gets worse and worse and after a minute, you're left with a dark raster and no sync Cause: It was a combination of Quartz and the 7404 driving the quartz. Had to rebuild the entire circuitry using another quartz and LS04. Symptom: Backgrounds rarely glitching to white now turned solid white Cause: A long-legged resistor was barely touching another metal component (or was it?) Symptom: Marquee lacking lights Cause: Shorted triac burnt trace. Qix Note: Standard procedure is first to replace the 2 resistors and the one transistor in the battery charging circuit. Symptom: Dead Cause: Battle on the Bus, no Reset. LEDs show 10100 Cause for no reset: Corroded connector pins but that was the last thing I found and fixed, reset worked intermittently, allowing me to continue repair. Cause: Both Video CPU and Main CPU have bad 2114 RAMs. One 4116 Vram dead too. Cause: Entire battery charging circuit rotted away. Had to replace 2 resistors, 1 transistor and a cap. Cause: CPU board LS245 get hot, replacing them ended the bus fights Cause: U100 (LS08) had one bad gate. Symptom: After Reset, LEDs show random values, randomly reacts to test switch, then LEDs randomly change Cause: LS138 next to video CPU bad. Cause (well not actually, but as this chip was clearly dead I swapped it anyway): one LS299 looked like an EPROM without window, had invalid outputs. Symptom: Unsynched bluescreen Cause: CSync chip (LS86) bad. Symptom: Colors weird. No Qix, some other graphics elements missing. Cause: Self test hung at binary 4, indicating color RAM problems. Symptom: No A8 on color RAM, but A7 and A9 fine Cause: Bad LS157 (U55) Symptom: Now there's CGA snow, characters are missing half of themselves, crashes in attract mode at completing the first line. Also draws small white rectangles that look like a small version of TARG. Cause (for the crashing and the distorted characters): LS244 next to the video CPU had a bad output. Symptom: Still CGA snow, still weird colors. Qix is now there, but now the Stix as well as all the fillings are missing. The fuse, too. But the marker and the Sparx are still there. Self test passes, but VRAM test only shows 32 colors instead of 256. White text has weird artifacts. Blue Qix has weird artifacts. Cause: Another LS157 (LS104) dead. Symptom: Colors normal, playable, but still CGA snow. Cause: Another LS299 (U110) had the weird failure mode of putting signals out of its INPUTS! (machine was now completely working - but not for long) Symptom: Erratic yellow pixels Cause: Bad Video DRAM Symptom: Only writes to half the screen Cause: Bad LS139 (U87) Symptom: Frequent crashes, object drawing is intermittent, looks like an intermittent contact Cause: U103 (LS00) bad. Symptom: Colorful stripes in some colors, game runs, self test passes Cause: Another LS299 Symptom: White stripes accross the screen, game runs, self test passes Cause: Yet another LS299 (that's four now) Symptom: Dead again. Sometimes LEDs turn off, sometimes they stay in the pattern they are before the board is reset on powerup. Sometimes they light up and turn off when you press test. Cause: Bad Battery SRAM Symptom: Lots of erratic white pixels, looks like Video DRAM Cause: you guessed it Symptom: Screen mirrored 4 times horizontally (that is vertically since the picture tube is in portrait orientation) Cause: Intermittent. Problem went away after fidgeting on every connection. With the problem there the self test still passes and since the memory locations differ, it can really be only in the video display circuit (that narrows it down to about 10-20 chips). Machine worked fine next open saturday and the next few after that one without killing a single IC. (one year passes) Symptom: Dead. Self test stops at 4 (color RAM) if it even gets that far Part of the cause: Ribbon cable connectors touchy. Resoldering them fixed the connection reliability issues. I have no clue how to put this in a short format, but testing all ICs passed, but when I hooked up the ROM board, the FI fuse blew and afterwards the CPU got real hot and I tested the CPU and the CRTC, both were bad as were two LS244 next to the CPU, swapped them and everything works again. Symptom: Blank screen, all LEDs flash in sync Cause: Bad pullup resistor network Symptom: Self test hangs at 50% during VRAM test Cause: The last of the four LS244s next to the video CPU died. Qix (2nd boardset) Since now I have a working boardset, I can split this into boards. ROM/IO Symptom: Dead. Self test says bad QQ9 Cause: Battery urinated on QQ9's socket. Replaced socket, game runs. CPU Symptom: Dead. Self test inconclusive. Cause: Bad 6850 RS232 controller Video Symptom: Blue screen with Memory error (colors don't match). Self test fails in Video RAM so color RAM OK Cause: U103 (LS157) Symptom: Just died with the same weird LED Patterns as #1 when it showed first signs of live Cause: U100 (LS08) Symptom: VRAM test fails Cause: A total of 4 4116 DRAM chips were bad Symptom: Color stripes accross the picture Cause: Good ol' LS299. Surprised? Symptom (after a year of lying around): Lots of colors missing Cause: It appears one of the LS299 is responsible for most of the colors... Qix (other version with 2 boards) NOTE: There are multiple versions of the 2-board Qix with different pinout. Symptom: Black screen (but it initializes the CRTC correctly!) Cause: Marginal/incorrect battery RAM consumes a lot of power when switching, glitching its own power line, causing the crash. Remedy: Short resistor on the battery power bus (or install battery...) Qix (another version with 2 boards and different pinout) Symptom: No video even though self test passes Cause: Transistor Q1 broke a leg Slither (uses Qix video hardware and very similar CPU board layout) NOTE: To get the board running on the bench, you must -Pull reset high, then low -Bridge the cashbox switch (a 2 pin connector below the video output) -Push the self test switch BEFORE it goes to audits (which is a window of approximatively 3 seconds after power on) Symptom: Fails self test with ROM error Cause: Bad ROM. Symptom: Fails manual self test with video RAM error Cause: Just a few dozen bits in one of the 32 video RAMs. Astrocade (GORF/Space Zap): Symptom: No picture Cause: Bad C2860 custom chip (in two board sets!) Symptom: No picture Cause: Shorted caps in video ram board Symptom: Dead pixels on some video ram boards Cause: Bad RAM ICs Symptom: Most the picture unrecognizable Cause: If only one of the RAMs on one of the board completely dies, this happens Symptom: Entire screen is static, but doesn't change Cause: Appears that the 12V capacitor on the I/O module has shorted just after the Data Custom has initialized. Symptom: Quickly fading green screen Cause: Said shorted cap on *both* I/O boards Symptom: GORF runs now, but: No sound Cause: Capacitor on sound board shorted Cause: Power supply fuse blown because of shorted cap Symptom: Ship sometimes moves right by itself Cause: Marginal LM339 on joystick board Symptom: Can't fire Cause: Bad trace on joystick board Symptom: Space General doesn't light up Cause: Bad bulb socket (bulb was ok) Symptom: Screen arcing somewhere at the neck Cause: Unknown. I thought it was the tube since the problem went away with a new CRT, but it didn't. It just works most of the time now. Symptom: Random crackling noise on right channel after about 10 minutes Symptom: No more sound on right channel Cause: The I/O custom IC that does the right channel sound has died Symptom: (spare board set) Starfield looks weird Cause: Bad LS161 on I/O board Symptom: Erratic yellow pixels in GALAXIANS when the enemies in move in formation (only on their line) Cause: Bad ROM (you read right, not RAM). Swapped ROM board, problem gone. NOTE: (when it was "working" self test showed a small x in the bottom left corner, after the glitches appeared, self test would show bad RAM, but it wasn't the RAM) Real cause: Bad sockets. The ROMs all verify OK except the German ROM where BEDNDET instead of BEENDET seems to be the only fault. Crazy Kong Symptom: Dead/Flaky Cause: Bad ROM sockets. One machine had a bad ROM where there was a scratch in the EPROMs window leading to bits flipping back. Vanguard Symptom: Sometimes speech, sometimes not. Noisy audio all the time. Cause: Bad tantalum caps on the speech board Scramble Bootleg Symptom: Dead Cause: Dead main CPU Symptom: every second column repeats the last two (like Galaxian back then) Cause: Probably a shorted trace. Problem never showed up on the bench, could make it go away entirely without really knowing how. Symptom: No engine noise Cause: Bad 4066 7H (the leftmost) Scramble Bootleg (another one) Symptom: Dead Cause: Sound CPU!? (the main board should be completely sound board agnostic) Symptom: Dead with short traces of garbage flickering up the screen Cause: Bad Flipflop @9L generating the CPU NMI signal Symptom: No shell, but game plays OK, so it's just a visibility problem Cause: LS161@10J had no ripple carry out, so the next one in line didn't count Symptom: No sound Cause: Actually, this one had LOTS of bad chips on the soundboard. The entire Sound CPU bus looks ugly. They took 5 LS08 and abused them as bus drivers. 3 of them were bad, some even had low impedance inputs. Also the data bus transceiver 82T245 (=LS245) had no output on one pin, swapped it, signal came back. Still no sound. Searching EVERY chip for bad outputs I found a bad 7402. Still no sound but at least the BUS now has the same patterns as the one from the other sound board. Symptom: Found out that no interrupt was fired when a sound should play. Cause: Dead LS74 responsible for Interrupt (there are banks of LS74 for audio filtering, it's not one of these) Symptom: Position lights of player rocket very dim, turn on after a few minutes Cause: LS273 driving the color PROM out of spec. Symptom: Weird sound effects when cold Cause: Slave AY-8910 bad. Amidar Bootleg using the same hardware as the above Scramble Symptom: Dead with garbage screen. Cause: Well, it wasn't just the CPU. Symptom: CPU reset line pulses randomly Cause: Not a single bad chip, *LOTS* of them. As before I searched for bad chips and found LS367@3A, two LS161 next to the shell circuit, a couple of LS74 spread throughout the board and for good measure and repeating the first two columns all over the screen an LS283. Before I can go back to Symptom, I have to wait for spares to arrive. In order to continue debugging, I had to undo the hacks done to this boardset, now it is identical to the Scramble one above; it now "runs" Scramble. It looks scrambled (heh heh), but the game runs. Symptom: After changing the character ROMs to Scramble to assess the graphic glitches, the picture now only shows colorful vertical (that is horizontal in portrait mode) lines. Cause: No R/W signal to Color RAMs, shorted to +5V. Cause: Unknown. Shorting it to ground made the problem go away and all 5 color RAMs seem to work. Symptom: All characters repeat their left half Cause: The other LS283 has one bad output. Good thing I ordered two. Amidar Sound Board Symptom: Game board won't boot with this sound board attached. Undid all the hacks to the sound board. Symptom: You guessed it, no sound. But game now works OK. Cause: Just one bad LS08 this time, but the one LS90 was bad as was the LS28 which again I changed for an LS02, also an LS32 next to the bad LS90 (A3 or 4) Redid parts of the soundboard hack to give it the address space it needs to run Amidar Sound. It works. Since the mainboard has a Big ROM Hack with two selectable Games (Amidar and Scramble), the soundboard needed that too, but: Symptom: Again no sound Cause: Bad LS04 next to the EX-28 now LS02. Symptom: The entire boardset sporadically resets Cause: My external game change/reset circuit killed the existing reset circuit Fix: Cut pin 4 of the LS00 that was connected to pin 5 which got the reset signal straight from the Power On Reset Capacitor/Transistor circuit and hooked it up to the game change button. Symptom: No more sound Cause: Another LS08 from the known batch died Symptom: No more sound Cause: You guessed it. One more to go. Symptom: No more +5V Cause: Unknown. I swapped the entire silicon in the +5V regulator with no effect. The resistors and capacitors all check out OK. I settled on leaving the current sense pin floating. Symptom: No more picture, no sync Cause: The very first horizontal LS161 died. Symptom: Back to the garbage screen Cause: The LS367 I salvaged from an old unknown and battery acid corroded board had died. Symptom: No sound Cause: Probably one of the AY Sound chips. Swapping the chips made the sound go slightly weird, swapping them back muted the machine. Symptom: Unsynched mess on the screen Cause: Next LS161 in line (horiz. 2) failed Amidar (Super Cobra Hardware) Symptom: Vertically (if 4:3) every 4th character gets repeated Cause: LS157 @ 5K (below the CHR RAM) had an input that pulled down the LS273's output. Symptom: Crashes when trying to draw sprites, or even not booting Cause: Bad ROM Revolution X: Symptom: Self test proclaims U67 and U77 (two SRAMs combined to make a 16 bit one) to be faulty Cause: The bus driving the bus driving the SRAMs had no connection on DB0 due to leaky capacitors rotting away the trace. (weird that this was the only bad trace, the machine worked right after fixing the connection) Symptom: Gun 1 no recoil Cause: Blown fuse Symptom: Funky colors/dragging pixels/video crazyness when cold Cause: /OE signal to color RAM address latches missing due to corroded trace Mortal Kombat: Symptom: Ingame: A few graphics are missing, noticeably the story mode info screen, a few backgrounds are glitching, but barely noticeable, plays great. Symptom: Self test red-flags EVERY single graphics ROM in that huge bank of graphics ROMs Cause: Probably the FPGA generating the chip select for those ICs. Tutankham bootleg Said to be very rare, the guy who brought it said that he didn't find anything on the internet about his version and the ROMs match up to nothing Symptom: Said he only gets colorful lines on the screen Symptom: Just graphics garbage, but the VRAMs are OK apparently Cause: The machine I tested in had the -5V line missing on the JAMMA Adapter Symptom: Colorful stripes with LSD syndrome Cause: Probably bad contacts to the daughterboard which basically rebuilds Konami custom chips into 7400 logic. I could just push on anything until the picture was perfect. Replacing the single wipe sockets to the daughterboard with something made for the pin connection used by the daughterboard (aka the Wipe-Destroyer aka the Wont-fit-in-machined-sockets) fixed the problems as well as making separating the daughterboard from the mainboard next to impossible... Pub Time Dart machine (Merit DRT-950) Symptom: No picture Cause: Bad CRT yoke, CRT also near-dead (needs regeneration) Cause: TMS9129 not generating any video signal Cause: Video CPU stuck in a loop Cause: The battery in the 8K SRAM wasn't the only thing broken in this chip Symptom: TMS9129 not generating any video signal Cause: Video output buffer transistor shorts signal to +5V. Had I fixed that earlier, I would have gotten the message BATTERY SRAM BAD on the screen. Asteroids: These two boards come from someone else. Symptom: Dead, RESET stuck low Cause: Watchdog LS393 last output dead Symptom: Binary Z-axis game plays with only the ship on screen which sometimes looks jumbled. Remaining lives set at an upward angle. Self-test missing gradient and Jumper position display Cause: LS32 driving the LS273 driving the Z-axis as well as some Y-axis bits had one bad output. Next machine (still Asteroids): Symptom: Dead, watchdogging Cause: CPU Work RAM. Symptom: Still dead, but self test proclaims Vector RAM bad Cause: Vector RAM bad Symptom: No change after vector RAM swap Cause: 2 bad LS157 in row 2 connecting the vector CPU to the 6502 Symptom: Machine attempts to draw the diamond pattern in self test, but only faint garbage on screen coming from stuff deflected way off the screen, bips constantly (high-pitched short beep) Measure: Looked at ROMs, still original, removing them also removed some of their pins, they just rusted off. Cloned the Vector ROM from the other machine. Part of Cause: The 4016s in the deflection unit. Symptom: Draws gradient, attempts to draw diamond pattern, crashes, repeat. Symptom after taking board home: Dead. No Z-axis (can't test X/Y on board, but vector state machine appears dead) Cause: LS74 at A7 outputs were pulsing but did not match the expected behavior Symptom: Machine takes some time to start before turning on Cause: LS04 next to Quartz Cause: LS393 at A3 (or near there) were both marginal and only worked when warm Measure: Removed L6 to be able to lift Pin 1 to facilitate troubleshooting Symptom: Watchdogging even in test mode Cause: The three CPU ROMs also fell apart when I removed them Symptom: Vector CPU still not running straight Cause: Randomly looking at ICs I found a defective LS86@ (P5-ish), one of the LS367 was bad, as was an LS83 full adder. Symptom: Still watchdogging in game mode, no Z-axis, but self test proclaims again that it can't access the vector RAM Cause: Another LS157 in row 2 died. Symptom: Game runs with one asteroid moving down blazingly fast, objects don't move on their Y-axis at all, but teleporting works. Self test fails ROM test. Cause: The ROMs I burned could not keep up with the 1.5MHz Clock speed. Symptom: Missing saucer shot sound, saucer sound is always for the small saucer even if the big one is on screen Cause: The only LS259 on board (near the audio section) Yet another Asteroids Symptom: Watchdogging. Selftest does one low beep. Cause: The main CPU RAM was completely dead. Symptom: No picture even though Z-axis looks perfect now Cause: Dead DAC deflected the picture completely off screen, replacing it shows the machine is visually working fine Symptom: Lots of sounds missing/corrupt/garbled Cause: Garbled audio and missing thrust sound were due to the LM324@P11. Symptom: Both fire sounds not decreasing in frequency Cause: The other LM324 was from the same batch and just as dead. Symptom: Stairs in both axis (looks like jailbars) Cause: BOTH DACs bad and my AD7533JNZ adapter wasn't quite finished And another Asteroids Symptom: Dead, Work RAM never gets selected Cause: Bad 74139 (the one closer to the CPU) Symptom: Test says bad Work RAM Cause: Half of it was bad Symptom: Test says bad Vector RAM Cause: Half of it was bad too Symptom: No Z axis, X axis sometimes stops working Cause: K7 had a bad blanking output Symptom: Still no Z axis Cause: No BVLD, reason for that: Bad 74191 (C or D) Symptom: Lines are squiggly Cause: Another 74191 (at around F or something), it passed the test, replacing it fixed the game though Symptom: No thrust sound or asteroid explosion sounds Cause: Bad 7486 in noise circuit (game works now) Symptom: One week later, completely dead, no watchdogging Cause: Bad 7474 in watchdog circuit Symptom: No game, no self test, watchdogs slower in test Cause: (I'm not good at the Vector CPU so I just shotgun chips with the In-Circuit-Chip-Tester) Bad 7402, bad 74273 (the rightmost one) Symptom: Starts drawing a screen, then crashes and resets Cause: Bad 7474 at A(8?) Symptom: Some of the text is garbled, starts counting at 1 instead of 0, half the asteroids and UFOs are invisible. Cause: Turns out the Vector ROM actually still works and is incompatible with the other version Symptom: Everything that moves wraps around every other inch Cause: Bad 74157 right next to the Y-74374 Symptom: Can't set it to free play - DIP switch bit 7 is stuck 1 Cause: Bad 74253 next to DIP switch Symptom: X axis bad, appears to be missing a bit Cause: Bad 74157 next to the X-74374 Symptom: X axis bad, appears to be missing a bit (again) Cause: Bad 74157 next to the one I just replaced. Symptom: X axis has some slight squigglies on it, looks like an analog problem Cause: Nope, just another 74157 in that row. I think chip 5 from the right. Tempest Symptom: Picture looks like garbage Cause: Bad Y-DAC Symptom: No scaling effects on the title screen (bad DAC08) Cause: Bad cap on pin 16 of DAC08 Hang On Symptom: Garbage screen. Test flags every RAM and ROM on the secondary CPU bad Cause: Bad LS157 in a bank of LS157s between both CPUs. Symptom: Garbage screen. Test flags one single RAM on secondary CPU bad Cause: Self test was right, that RAM was bad. Symptom: Renders half a road and crashes before even drawing a foreground layer let alone sprites Cause: (that took forever) The other RAM on the secondary CPU which passed the test was also bad! Symptom: Garbled sprites that repeat in X-direction and are mega glitchy Cause: Bad LS194 in sprite RAM circuitry Super Hang On Symptom: Sprites mega corrupt (only a few single lines), one foreground layer missing (CRT test doesn't show the gradients, the screen is nearly completely empty) Cause: Probably one of the big custom ICs. Sadly this is a death sentence. (also Cause: One of the sprite RAMs was bad, but changing it didn't do much) Super Hang On Bootleg Symptom: Super flaky with random colorful lines. Sometimes INSERT COINS flashes on top of that garbage, but rarely. Cause: The board is a mess (eg. I found an LS138 where they forgot to tie the enable pin to +5V, there are HC356s in the graphics section where only 18 pins are in the board and the two extra pins - CLOCK and GND - have been cut off all the chips - this makes no sense, but I was told this board once worked). Cause: One 8k RAM bad, several ROMs bad, although two have a different addressing scheme and can't be found in MAME. Symptom: Flickering Garbage Screen. Cause: Can't get the 5V high enough. The board works "fine" on a PC power supply. (fine for this mess of a board) (using the MAME ROMs instead of the ones that didn't match on this board) DECO cassette system (this is not going to be about the cassette) Symptom: Jailbars Cause: Bad video RAM (they're on the board with the sound CPU/AY-3-8910s) Symptom: Wobbly picture Cause: Dry joints in power supply Symptom: After a week, wobble came back double strength Cause: "small" value capacitors in the primary bad Symptom: Still a bit wobbly, but much better now. Swapping the third cap made it stop wobbling once. ---I got a Darksoft Cassette Replacement Multikit--- Symptom: Horizontal sync starts glitching after an hour and the machine randomly crashes Cause: The power supply lets the 5V dip when it gets warm. I replaced the PSU. Symptom: Randomly appearing Jailbars, games randomly crash Cause: Another video RAM started dying Symptom: Multikit doesn't boot reliably Cause: 47pF deglitching cap necessary for early board revision measured 42pF (replaced it with 56pF, now it's solid) Symptom: Machine froze, garbage screen after power cycle Cause: Bad 555 in power up reset circuit Another DECO boardset Let's call the boards Glue, Sound and Bios (the latter having the epoxy block 6502 CPU) Glue (actually called BIO): Symptom: Crashes when trying to load game Cause: Two of the RAM chips were bad. That's all that was wrong with it. Bios (called DSP): Symptom: Garbage screen Cause: One LS10 in the corner was super rusty and also pretty damn dead. Symptom: No shots in Super Astro Fighter Cause: Bad LS74 in the row with the quartz. Symptom: Some games crash, most games run too slow, some run too fast Cause: Bodge IC (74LS00) next to clock IC not populated Symptom: Now it doesn't react to the start button any more and starts behaving randomly with random crashes and glitches. Cause: Hold the line... at least one of the sockets was bad. Cause: For some reason it only works fine with its original RMS board. Cause: Get that, if I swap positions of RMS and DSP board, it works fine Symptom: A few scrolling games have background tilemap corruption Cause: Unknown, but it goes away when the machine runs for a couple of minutes. Sound (RMS): Symptom: No sound Cause: No clock to 6502 CPU Cause: A bad LS74 in CPU clock circuit. Symptom: With the Multikit, the machine renders one initial frame of garbage and freezes Cause: There was a bad LS367 in the dongle circuit, but that didn't fix it and now every single chip on the board is tested working and the problem persists. Cause: There was another 74367 next to the edge connector that had the ground pin bent over from the factory, but of course that didn't fix anything. Cause (finally): A bad RAM chip that had addressing faults, so clearing/ setting didn't reveal it. Symptom: Electronic interference noise Cause: Bad 555 Symptom: Non-stop beeping from the 555 Cause: Reset signal coming from the AY-3-8910 sound chip missing Cause: Bad AY sound chip Cook Race (Burger Time bootleg) Symptom: Glitchy graphics on one layer Cause: Cracked trace. I can't find it, but if I bend the board just right, it works fine. One of the spacers for mounting the board is longer than the others and if I were to mount it in a machine, it would work. Bomb Jack Symptom: Audio channels missing (sounds thin) Cause: Completely open capacitors Symptom (different machine): Title screen and sprite layer has some faint glitches on every 16th scanline Cause: Bad power connections ---Next boardset--- Symptom: Foreground and text layer completely messed up Cause: RAM 4 bad (error message barely legible, had to look up the position on a working machine) ---Next boardset--- Symptom: Background shares color with one of the other palette, leading to the background doing rainbows sometimes Cause: Unknown. The self test passes. In game it looks more like a feature than a bug ---two more boards--- Symptom: Passes selftest then just resets Cause: Bad cntacts on interconnect. Symptom (other boardset): Dead Cause: Same. Interchanging one board per stack made one machine come back and putting both boardsets back together fixed both) Phoenix (mine!) Symptom: Colors wrong, only 8 primary colors showing Cause: Both color PROMs were swapped. Symptom: Ship randomly moves right Cause: Dry joint in joystick board assembly Symptom: Shot/explosion sound gone (Noise generator dead) Cause: Bad 555 timer. Symptom: Buzz and ripple through the picture Cause: Bad rectifier ---Next boardset--- Symptom: Only 8 primary colors Cause: (maybe bad color ROM, maybe bad connections, maybe same as mine) Symptom: 4H signal bad on foreground character layer Cause: Cracked trace somewhere I can't find it, but bending the board helps ---Next boardset--- Symptom: Barrier not working Cause: Error in the bootleg traces, it's routed to a different pin. Rolling Thunder Symptom: Plays blind (no picture) Cause: LS273 in the blanking circuit completely dead Gauntlet Monitor (some Wells Gardner iirc) Symptom: Picture slightly too small and wobbly Cause: The small transistor in the voltage regulator circuit (BC237) Gauntlet Symptom: No speech Cause: Disabled in settings Symptom: Garbled speech Cause: Bodge capacitor generating the chip select signal for speech was bad Gauntlet (another machine) Symptom: Every 16 pixels, one line is missing certain colors Cause: One of the SLAG custom ICs has one input high impedance Moon Patrol (all bootlegs, all 11 of them. Most work though) Symptom: No sprites (they disintegrated after a second, now they're gone for good) Cause: An LS04 on the CPU board couldn't keep up with the timing of very high frequency signals (generated by an LS10) Symptom: No sound Cause: 384kHz quartz for explosion was broken off, doesn't explain why the sound didn't work in the first place, but now it works Symptom: Doesn't fire Cause: Both input LS244s had invalid signals on their outputs, looks like the harness once got some high voltages (12V or -5V) sent into it. ---next boardset--- Symptom: One of the colors looks inverted-ish Cause: Short circuit on a bodge wire near edge connector ---next boardset--- Symptom: One color is missing. Probing on the bench shows a very weak and fuzzy signal Cause: Pin broken on the interconnect. Zippy race aka Traverse USA (bootlegs, VERY similar to Moon Patrol) Symptom: CPU missing. With a new one, garbage screen that changes every second Cause: One ROM was missing. It was never populated, there was no socket in its position. The three ROMs in the machine were identical to the four-ROM version of the game. Kung Fu Master (almost same HW as Zippy Race) Symptom: ragged stripes on screen. They do not move ever and don't affect sprites Cause: Dry joints on a custom SMD Symptom: Jailbars in sprites Cause: There were remnants of one of the sprite RAMs still on the board... Turbo Outrun Symptom: No sound Cause: Couldn't find it, probably bad muting signal, removed the final LM324 and the sound now works. Blood Bros Symprom: No picture Cause: The custom IC generating the signal that goes to pin 6 on the PAL that generates blanking Remedy: Pull pin 6 of that PAL high. It will leave ~16 pixels of garbage below the screen, but at least the machine is back to playable. Symptom: Some (but not all) palette animations not working, some colors wonky Cause: Looking on the internet, there's another guy with similar problems that were also caused by that exact same custom. Green Beret (Bootleg, original 1985 Konami) Symptom: CPU on HALT Cause: Self test fails with ROM 3 BAD. Burning a replacement fixed the board. Jail Break (Bootleg, original 1986 Konami) Symptom: Garbage screen Cause: Pins pinching traces on solder side, pins shorting against each other Symptom: Jailbars in sprites Cause: Two out of four 8k SRAM chips bad Dribbling (Model Racing) Symptom: No Sync. But picture looks promising Cause: Buffer chip right before the edge connector has a bad output. (nothing else seemed to be bad, the game runs fine) Espial Symptom: Same. Picture, but no Sync. Cause: This time it was an LS08. Symptom: Weird sync, weird picture, less than 100 scan lines per frame Cause: Contacts. Flexing the board makes this problem come and go Symptom: Shows "RAM CHECK OK" then reboots instantly. Cause: ROM 4H has a few bits flipped back to 1 that should be 0. Symptom: Sometimes scrolls only part of the screen, some columns "hang", sometimes it inits with weird sync again Cause: Flaky interconnects COP 01 (Nichibutsu) Symptom: Stripes in some sprites and parts of title screen Cause: LS377 latch for the upper GFX ROM bank had two bad outputs Symptom: A few sprites have weird stripes in them Cause: Someone swapped the first upper and lower sprite ROMs. Popeye (bootleg) Symptom: Very glitchy graphics, including stripes in graphics Cause: A bent pin shorted an address line to a data line Symptom: Ragged stripes in sprites Cause: The ROM's sticker had a scratch whose shape actually matched the raggedness of the stripes in the sprites - this board was stored in sunlight. Symptom: Slightly weird blocky colors on the title screen Cause: All bootlegs do that Puck Man (probably bootleg) Symptom: Garbage screen Cause: Reset never goes LOW Cause: (unknown, the problem went away on its own) Symptom: Black screen with two Pac Men with stripes through them, then crashes Cause: Broken trace Nibbler (probably original) Symptom: Dead a week after swapping +5V and GND (D'OH!) Cause: Bad BUS transceiver at edge connector on video board Symptom: Still dead Cause: Bad IC sockets Elevator Action (Original) Symptom: White screen Cause: Custom CPU board (encryption fix) had the corner 4 pins soldered which broke off Symptom: Dead Cause: Can't produce it on the bench. It does look as if /IRQ is bad or missing, the game exhibits significant drops in the framerate. Cause: Probably a bad interconnect Symptom: Wonky blue Cause: A 7417 that wasn't on the schematics had one output bad. ---next boardset--- Symptom: Massive amounts of CPU data are snowing over the video signal Cause: Using the working boardset, I could eliminate the problem to the top board. The board appears to be incompatible, maybe someone switched it with a board from a different game. There was a bad LS374 next to the PAL (which was working and compatible). Symptom: It did work twice with minor flakyness and cocktail mode on Cause: 9 bit color RAM wasn't in right Symptom: Now it's dead again Cause: One of the LS193 on the ROM board died (very early OKI). Symptom: Graphics glitches Cause: Probably another LS193, swapping the ROM board fixed it. Symptom: No shot sound Cause: Bad AY-3-8910 (the middle of the three AYs next to the Z80) Jungle Hunt Symptom: Graphics use the wrong tilemap, text is completely white Cause: Bad LS193 on ROM daughterboard. Street Fight Symptom: No boot Cause: Bad crypto CPU Symptom: No coin (different board) Cause: Also bad crypto CPU Halley's Comet (Taito original) Symptom: Game visible behind blizzard of CPU data Cause: Two traces scratched through Symptom: No sound Cause: Two cracked potentiometers Another Halley's Comet Symptom: Dead. Two of the 32 video RAMs have no/bad output Cause: Stuck in reset. Replaced these bad RAMs. Symptom: Stuck in reset Cause: The huge board set has two disjoint 12V rails. Both must be connected for the overengineered Reset circuit to work. Space Cruiser (Taito original) Symptom: CPU data snowing over video image of the game running Cause: Flaky interconnects Jackson (Zaxxon bootleg) Symptom: No clock Cause: Bad Quartz Symptom: Clock does not get passed to rest of board Cause: Bad 10125 ECL chip Symptom: Sprites look like NES with too many sprites on screen Cause: runs off 4.2V instead of 5V. Symptom: Shot sound suddenly shortens Cause: Bad caps on sound board Future Spy (Zaxxon HW) Symptom: Garbage screen Cause: CPU socket bad Time Pilot bootleg Symptom: Graphic errors: Pixels have double the size (both x and y) Cause: Konami custom IC replacement board not seated properly ---next boardset--- Symptom: Massive sprite corruption Cause: Bad sprite RAM 1942 (Capcom) Symptom: Draws title screen with garbage sprites and crashes Cause: No /WE on sprite RAM due to bad LS112, also one LS08 had a bad gate Jackrabbit (Zaccaria) Symptom: Black screen (shows garbage, resets, clears garbage, crashes) Cause: No NMI from video board due to bad LS74 (located on board edge right at the gap between sound and CPU board) Symptom: Joystick not responding (all DIP switches, coin and start work) Cause: Operator error. Needs to hold diagonal and press fire to move. Le Bagnard (aka Bagman) Symptom: Multiple vertical issues (4V, 8V and 32V at least) Cause: Two LS283 bad. Super Bagman on Crazy Kong hardware Symptom: CPU noise in video (only backgrounds, not sprites) Cause: Bad LS245 on background RAM BUS (one above the video ROMs) Super Bagman on regular Bagman hardware Symptom: Sprites disappear partially or completely when they should go behind the background (very subtle problem) Cause: Golden Axe Symptom: Graphics errors on one of the character layers Cause: Bad RAM Tiger Heli Symptom: Barely any remnants of sprites Cause: LS669 next to RAM is hanging on two legs, must've been hit Symptom: No change? Cause: Two scratched traces Another Tiger Heli Symptom: Sync frequencies jacked up Cause: 74S04 next to quartz going out of spec And another Tiger Heli Symptom: Irregular garbage lines through sprites here and there Cause: Bad 8K SRAM @ J2/3/4 (that's one IC - and it was a Hitachi...) Cosmic Cruiser, Typhoon... Symptom: Sprites with Jailbars Cause: Scratched traces, pins bent so they short each other or traces... Hyper Olympic Symptom: Screen full of white Ns Cause: Deep discharged battery prevents RAM from getting power ---next board--- Symptom: Wandering garbage screen Cause: Scratched trace to program ROMs ---next board--- Symptom: Screen full of white Ns (with no battery) Cause: Self test fails, one bad 2114 RAM ---next board--- Symptom: Solid white screen (with ghosting of the game running behind) Cause: Bad LS375 right on the edge of the video board Symptom: Sprites missing every other line Cause: Two bad LS161 ---next board--- Symptom: Video Address Generator malfunctioning (aka distorted picture, weird sync) Cause: Two bad 74161 in video address generator (Hitachi!? must be counterfeit) Symptom: Screen full of white Ns Cause: Same 2114 RAM bad as two machines up Symptom: Sprites look like a thunderstorm Cause: Deglitching cap removed for replacing the 74161s wasn't put back ---next board--- State: Super filthy, corrosion everywhere Symprom: Is ded. Cause: Bad address generator custom Symptom: Just a bunch of garbage sprites Cause: Bad D5 program ROM (on position A3), bad address buffer (LS244, C4?) Symptom: Yellow screen, flickering garbage sprites. Cause: More bad ICs. Bad LS04 next to battery, three bad LS157s on the other side of the board, one bad LS244 (less important) Symptom: Yellow screen, random (but recognizable) sprites from the game unfortunately, the custom in the bottom right corner is dead as well so I give up. Silk Worm (?) Symptom: Dead Cause: HC04 in clock circuit bad Symptom: Black screen Cause: Sprite ROM super bad, trashes the BUS Do! Run Run Symptom: Barely any sprites Cause: Scratched trace between Z80 IORQ and custom IC Symptom: Sprites with jailbars Cause: Bad 4164 RAM Pento (Pengo bootleg) Overvolted big time, looks like the board was briefly on fire, incandecent traces... Symptom: Voltage protection, LS368 gets hot Cause: Several shorted ICs NOTE: Some ICs have inputs shorted to ground which can be found by pulling the signal to +5V then using the finger test to find the hot chip(s) Bad ICs were mostly in the vincinity of the IC that must have been on fire Many of the burned-through traces were stuck low. Pulling them to +5v revealed many bad ICs Symptom: No sync, no video Cause: Bad LS74 and LS161 in video address generator circuit Symptom: Traces of video but still no sync Cause: Bad LS157 next to blown up IC (LS258) Symptom: Garbled/no sync Cause: Second vertical LS161 has enable stuck low, comes from bad LS174 next to graphics ROM Symptom: Good sync, good garbage screen, CPU watchdogging Cause: Bad Z80, every single LS367, an LS377 and a bad LS138 Symptom: Boots to black screen Cause: LS253s used for inputs were bad. Symptom: Sprite positions flaky/messy/glitchy Cause: Corroded IC legs in sockets Symptom: Minor sparkling in sprites Cause: Don't tell anyone it's not a feature! Tapper Symptom: Blue base level too high, no black, some colors weird Cause: 82S09 64x9 bit RAM lost the middle blue output bit Pac Land Symptom: Garbled garbage screen (with weird sync) Cause: Cut trace Symptom: Error code 2 0 0 (RAM 9N, which is OK) Cause: Bad LS32@10L Skylancer Symptom: Sprite positions sometimes wrong, sprites avoid certain areas Cause: Bad LS245 in sprite circuit. Piranha Symptom: Boots to a black screen Symptom (after months of absence): The black screen develops a flickering grid of 0s that fade in and out like a bad trace. Cause: Bad 2114 RAM Symptom: Glitchy video Cause: Replacing another 2114 who fought against a piggyback solved the problem. The 2114 passed though and works perfectly fine in another game now. Toki (Juju Denetsu) Symptom: Missing 2 background layers, sprites consist of garbage Cause: Some bad video RAMs Symptom: Sprites missing every other scanline Cause: I ripped a couple of traces swapping the video RAM and missed one Symptom: Back to garbage sprites Cause: Bad interconnects. Jiggling them also brought back the missing background layers. Symptom: No sound Cause: ? measuring the amp I slipped, causing a short circuit which caused the sound to work so I made it a permanent short circuit. Nothing's overheating. Euro Leage (label says "Viesel Soccer") Symptom: Dead Cause: 3 out of 4 Toshiba 2k SRAMs were getting brutally hot and tested bad Symptom: Garbage screen (different every power cycle, sometimes the game boots and runs) Cause: Missing capacitor in Power On Reset circuit Street Fighter 2 (if you think this is an original, you gotta be stupid) Symptom: Weird graphics errors as if a counter wasn't counting right, maybe bad video RAM. Cause: Unknown. The game is missing a Program ROM though and another one doesn't match, problem persists Solution: While piggybacking and measuring, the problem fixed itself forever with no indication of what went right and what was wrong. Aero Fighters Symptom: Dead Cause: Power On Reset circuit was completely destroyed Goindol Symptom: Most of the sprites missing Cause: Bad LS374 in sprite circuit Future Flash Symptom: No sync, barely any video. Resetting the machine sometimes gives short blips of video Symptom: CPU stuck in IRQ, only writes half of its video RAM Cause: Haven't fixed it yet. It's a Z80 which has to acknowledge IRQs, that it doesn't happen points to the ROM being bad, but the ROMs all pass. Decoding? Bad RAM? Hairline cracks in the traces? Cause: Turns out, one of the 2114 on the CPU board was bad. Symptom: Only boots to a solid khaki screen (with ALL chips on both boards known good) Bosconian Symptom: Interconnects missing. Bought stuff to make new ones Symptom: Pretty much dead. Using the other Bosconian to test boards individually Video Board: Symptom: Sync signal looks terrible Cause: Maybe a short circuit in the adapter. Removing the bodge and putting it back solved the problem Symptom: Video has solid backgrounds instead of black Cause: Bad BOS1-8 custom :-( CPU Board: Symptom: RAM 1L error message Cause: The only RAM on the board was completely dead. Double Dragon Symptom: Blank screen, no vertical sync component Cause: A bad LS00 in the clock circuit to one of the LS161 counters. Pitfall 2 Symptom: Dead Cause: Missing PAL (the first one for the clock signals) Symptom: (with a PAL from Wonder Boy) Still dead (at least CSYNC now) Cause: Broken trace from ground to the main BUS buffers' enable signal Symptom: Scrambled clocking signals with PAL from the internet Cause: Probably incompatible. Works fine with a PAL from a Wonder Boy BL. Cure: Wonder Boy BL works with the PAL from the internet, so I swap chips. Tetris (Original) Symptom: Dead Cause: Bad IC sockets Symptom: Dies after 5 minutes Cause: Bad POKEY Symptom: Completely dead Cause: Bad SLAPSTIC Tetris (Bootleg) Symptom: Clears screen, then watchdogging Cause: Bad 74LS240 next to edge connector Cosmic Invaders (Universal) Symptom: Dead, garbage screen, no boot, HALTs Cause: Bad BUS transceiver (74244) at J8. Symptom: Sprites are at garbage positions and never move Cause: Bad 7402 at P4 in the sprite RAM /WE circuitry. Sprint2 (Atari/Kee) Symptom: Dead, but in a flaky way. Cause: Bad RAM on bit 1. Symptom: Dead, black screen, most all signals missing Cause: CPU clock input turned into a short - bad 6502 (but it's still working) DragonNinja (DECO) Symptom: Every 2nd scan line gray Cause: One bad RAM of the two that are 90ø rotated from all the other RAM. Symptom: Swapped character addresses (ABCD becomes BADC) Cause: some tiny bodge wire must've come off Symptom: Glitches up or goes to copyright screen as soon as gameplay starts Cause: Via too close to a trace, shorted. Carnival sound CPU Symptom: No music Cause: Bad 8035 CPU Symptom: Only one bass voice playing Cause: Bad 8253 timer IC. Street Smart Symptom: Bad pixels in some backgrounds and some sprites Cause: Bad video RAM Vendetta Symptom: Dead Cause: Bad CPU PAL Symptom: Bad colors Cause: Bad color RAM Symptom: Distorted samples Cause: Sample ROM bad Symptom: Works fine, but motorbikes are square waves Cause Mixed up upper banks on double size EPROM. Shorting pin 42 to pin 1 fixed it. Monster Bash (Sega) Symptom: Wave sounds only work sometimes Cause: Bad Power Up Reset capacitor next to sound CPU Symptom: Scambled backgrounds Cause: Bad LS161 near the edge (two lone 161s next to a sync PROM) Frisky Tom Symptom: Stuck in RAM/ROM test Cause: It must be the board itself, all components pass or work when swapped with ones from an identical board. The fault is on the CPU board, not on the video board. Remediation: Short the 555, now it'll run. Silent Scope (#1) Symptom: Watchdogging Cause: Many damaged traces from someone desoldering the RTC chip Symptom: Hardware Error P11 (message on screen) Cause: Fixing the RTC chips by drilling them open and connecting a new battery destroys them. Silent Scope (#2) Symptom: Freezes in Ch (POST code) Cause: Silent Scope (#3) Symptom: Freezes in (I forgot, but before Ch, because video initializes during Ch) Schneider Euro PC (8088, 512k RAM) Symptom: Acid damage from the battery. Only gives a 10 second beep. Cause: One LS245 support chip had DIR and a data bit floating due to eaten traces. Symptom: Now one long and one short beep, still dead. Cause: JIM 5010 system controller had two data bits eaten away. Symptom: System is now alive, no RTC. Cause: The RTC is where the battery sat. You guessed it. Fixing all the traces put the machine back in business. Except for the parallel and serial port, but I didn't test them and they were reported as present in BIOS boot screen. Nordmende Film Scanner (Super 8), model 451a, 1975 Symptom: Very jittery picture, white line on bottom/top? of screen Cause: Wrong replacement transistor in vertical PNP/NPN pair Symptom: Not syncing to perforation, machine stops after a second Cause: Signals from light switch too weak, 1MegOhm pullup fixes that Symptom: Quit syncing to perforation, runs way too fast, capstan motor starts immediately on powerup instead of delayed Cause: Bad æA709 op-amp in capstan sync circuit ---ANOTHER almost identical Film Scanner 451a, slightly later revision--- Symptom: Color missing Cause: One of the RGB amps was bad, a tantalum cap was slightly leaky (a few hundred æAmps) which turned off one of the preamp transistors Symptom: Picture goes completely dark, partially inverts Cause: R-Y/B-Y matrix had one preamp transistor partially short to 24V (always good to have two identical circuits next to each other so you can see on the scope if something's amiss) Symptom: Hum bars (reddish / bluish) across picture Cause: The machine's optics are looking right into my ceiling fluorescent fixture. Turning the lights off fixed that Symptom: Playback stops after a few seconds Cause: Winding reel light sensor yellowed with age and got less sensitive Remedy: Replaced light bulb with white LED Symptom: Massive horizontal jitter Cause: Turning the CRT yoke fixed that IME 84 table calculator (1964) Symptom: Some keys don't work Cause: Bad keyboard contacts Symptom: Numbers missing, 7s turn to 5s, 2^1 bit missing Cause: Bad transistor in BCD to Nixie decoder Symptom (after working correctly for a short time) Symptom: Numbers missing, 7s turn to 3s, 2^2 bit missing Cause: Another bad transistor in the same decoder Symptom: Digit no. 5 not striking Cause: Bad transistor in column driver circuit Symptom: Numbers enter as 0xF Cause: Bad edge connector contacts Here some Vectrices (what's the plural for Vectrex?): Symptom: Funny picture. No text, just lines. Turns out blanking signal was missing. Cause: Pin 19 on the big CIA chip had no output. Symptom: No picture. Also the case broke during shipping and so did the CRT. Cause: That LF423 (I think) quad Op-Amp next to the XYZ-Connector. NOTE: If in a hurry and no spare parts, you can actually substitute this chip with an LM324 which are common as dirt and the fact that the XY deflection can only handle audio frequencies, you won't notice the LM324's smaller bandwidth. With a broken case and CRT in shards, this Vectrex, together with a broken IBM 5151 MDA monitor became the Wrectrex you can see on my site as well as Youtube. Symptom: Like the first, no blanking, but the Z axis exists, rendering the text legible. Cause: Had too much Oomph on the brightness pot. Just had to reduce the brightness. Symptom: Doesn't turn off. Cause: The lubricant in the power switch went stiff. Symptom: If you turn it on, nothing happens for half an hour, then it turns on. Cause: Same as the last one. If you don't like having to wait for the machine to come on or leaving it turned on and pulling the plug, you can power cycle it rapidly, getting the lubricant flowing again. But that just lasts a few days or hours. Cleaning the switch requires disassembly and (tricky) reassembly but can be done within an hour. Symptom: Dead. No sound, no vector chattering. Cause: The CPU was bad, but sadly not entirely, the BUS was still rather active, so I desoldered the ROM too in case that one was bad. Symptom: Now we have sound, but no picture. Cause: The XYZ-connector cable was bad. Symptom: Very weird picture. It has blanking and Z-axis, but geometries don't match anything. Cause: Bad DAC. Symptom: All graphics consist of a single diagonal line. Text is centered around that diagonal line, but legible. Cause: CD4052 X/Y multiplexer Symptom: Weak spot between lower left corner and center is the only thing that appears on the CRT Cause: Reseated CIA and that fixed the machine. Symptom: Plays blind, no picture, no buzz Cause: One side of the power switch worked, the other was gummed up like all the others Symptom: No picture, no sound (except pot scratching), but neck glow Cause: Bad CIA Symptom: No picture even though vector buzzing Cause: Bad DAC, bad 4052. Here some Colecovisions: Two french Colecos with SCART RGB. Symptom: Everything red appears purple, weird problems with yellow. Problem starts after 5 to 10 minutes, I was told. Cause: None. Could not reproduce. Works fine on the TVs I tested it on. The next one was found in a barn with a broken roof. The RF shield was a lump of rust. Symptom: Dead. If you turn it off and back on, it'll produce a black screen, but before that, just nothing, not even sync. Cause: One of the 2114 work RAMs was defective. Could tell because it would blank the screen, then crash and keep trying to write to RAM. PAL Colecovision Symptom: Weak and hummy RF signal Cause: Short circuit on the 12V rail. Cause: Using a milliohmmeter I found that the Video RAM on the right next to the GPU was the most probable cause. Symptom: Still not working, the Video RAM chip above the defective one gets brutally hot Cause: Well duh. Replacing both chips brought the machine back to working condition - almost. Symptom: Can't move down in Donkey Kong Cause: The pins from the joystick port were too long and two had shorted together. NOTE: If you build your own joysticks, note that the diodes in the switch matrix schematic are all the wrong way round. Those outside the matrix are correct though. Creativision aka Dick Smith Wizzard Symptom: Dead Cause: Bad 6502 CPU Symptom: Some keys on left joystick don't work Cause: Probably corrosion, taking the controller apart and putting it back together solved the issue Symptom: INPUT statement in BASIC is flaky, doesn't work for strings Cause: One bit in one of the OTP ROMs failed. Programmed it over, works. Interton Video 2000 (has nothing to do with the Video 2000 VCR format, this is a PONG home console) This is a discrete (discreet?) Super-PONG console with cartridges. The system has the paddle and video address logic and the cartridges have the game logic, all using CMOS 4000 series ICs in analog mode, greatly simplifying the paddle/ball logic. Symptom: No paddles, no ball Cause: Two 4069s failed. That's all, works fine now. Atari Jaguar Symptom: Not reacting to joypad inputs Cause: Bad HC244 shorting the enable levels to +5V Fairchild Channel F Symptom: Sometimes garbage in the middle of the screen, the right half of the screen is doubled Cause: A 74153 with a mishandled input pin lost connection to the bond wire. Soccer Kings (pinball machine) Symptom: Randomly deactivates and reactivates flipper paddles ingame Cause: Bad 2101 dual port RAM. Genesis (Gottlieb) Symptom: Thumper bumpers give no points Cause: Something fishy in the switch matrix: If you set some of the DIP switches, the idle high drops below logic high threshold. Cause: Probably what follows a year later: Cause: LS32 in switch matrix had outputs shorted to ground and super low impedance inputs, screwing everything that tried to drive them. Robot Wars (Gottlieb) Worked, but the Flasher next to the Spinner was shorted to the Spinner, shooting several dozen Volts into the switch matrix. That killed: The LS32 and two LS00 and the RIOT controlling them. (also the diode for the spinner) Flash Symptom: Temperature sensitive, crashes when too cold or too warm Cause: Bad 5V capacitor Odin (rare pinball machine, this one is actually a prototype) Symptom: Dead, starts doing weird things if you leave it on for more than 15 minutes Cause: Bad Z80 CPU Symptom: Nothing but watchdogging Cause: yet unknown, replacing the ROMs solved the problem* *) turned out to be MANY bad traces Symptom: Still dead Cause: Replaced dead Z80A with 1979 Z80 which can't run at 4 MHz. Symptom: No coils, no reaction to contacts, credit and start buttons work, no displays/displays only rarely get strobed Cause: Clock trace to Sound/IO Board corroded Symptom: Sound now works, some displays work, still no coils, doesn't count points Cause: Coil driver controller IC mapped to different memory location than series model (this is a prototype) Fix: Put original ROMs back in Symptom: Right flipper finger stays up Cause: Leaf switch contact too long, bends when playfield is lowered Symptom: Score displays don't work Cause: Unused dip switch (remember the prototype DEBUG ROMs?) not set to off Symptom: Worked fine for a year, now it doesn't boot anymore at all, not even the coil test. Cause: One of the two Program ROMs gets very very hot. Dumping it yielded some differences to its original dump from a year ago, burnt a new one, machine now works. Foxy Lady (Cocktail table pinball machine) Symptom: Dead Cause: had a bad acid trip Symptom: Reset not working Cause: Transistors eaten through, replaced them Symptom: Still dead Cause: ?100k resistor where 4.7k should have been? machine works now. Space Train Mac Symptom: No reaction to anything, but attract mode works, no displays Cause: Bad chip on the display board shorted two of the switch matrix lines Removed the display board, the machine now coins up and starts a game Symptom: hitting certain targets resets the machine Cause: Some idiot, probably me, put the switch matrix plugs on upside down. Symptom: Machine freezes during attract mode, always at more or less the same point independantly of which mainboard, driver board or ROM version I use Cause: Probably a short in the lamp matrix. Sadly, no. It also crashes when you don't connect it to anything. Also the ROM's contents hasn't changed from its original dump back when it was still "working" Cause: Coil power fuse blown Symptom: "Gives" four more balls and then goes to Game Over once the first ball enters the playfield Cause: As yet unknown, but all playfield switches are literally falling apart, I bet there are shorts and also probably a few bad diodes. Symptom: Bad logic signals going to the display board Cause: AY-3-8910 with bad I/O lines Symptom: Shows "ERROR2", no coil voltage even though everything is alright with the coil voltage rail Cause: Unknown. Grounding the AY pin "fixed" the problem Cause 1 for weird behaviour: The operating ROMs aren't as compatible as expected, but swapping it led to the discovery of Cause 2 for weird behaviour: The tilt pendulum was adjusted in a way that made almost permanent contact. This was essentially a short in the switch matrix that would clear from time to time but confused the CPU. The machine now works for the most part. Symptom: Left slingshot dead, right slingshot stuck on (and burns coil if left on for too long) Cause: Other AY-3-8910 had these two I/Os bad as well The machine now works 100% Symptom: Frozen, doesn't boot. Cause: Voltage leakage between the 12V rail and the AC test in used by the machine to derive the animation clock - the machine was waiting for a zero crossing that never happened. Cause for that: Capacitor leaked electrolyte. Note that the coil voltage detect circuit also resumed working. Symptom: After testing my reverse-engineered orange CPU board PAL, the machine is dead. Cause: The original PAL (which was actually flaky before) decided to die. Thankfully my reverse-engineered PAL worked first time. Space Train Mac Spare board Symptom: Dead Cause: Previous owner substituted dead 7805 voltage regulator with a 3.3V one Cause: One of the two ROMs dead Cause: Battery SRAM dead (I don't mean the battery - the entire chip is bad) Cause: Display/switch matrix controller chip bad Cause: LS04 in clock circuit bad Swapping the RAM for a 6116 one and putting in the ROM from the other board makes this board behave exactly the same as the other one (when it was giving ERROR2) Symptom: No 12V Cause: Uses different circuitry to activate the 12V - the other board just has a diode in place. Recreating the circuit on this board made it work as well. Symptom: No coils - doesn't engage GAME relais Cause: The transistor driving it was shot Symptom: No voice samples Cause: They're not getting triggered - one of the AYs I've swapstolen (swapped ones from machines where the I/O isn't needed) obviously isn't working 100% NOTE: I finally reverse-engineered the PALs (even the CPU one in case one dies and fixed one of the original display boards Symptom: All displays show "8"s. Cause: Marginal UDN6118 (they work fine if you clock them slow) Last Lap (Playmatic) Symptom: Won't boot Cause: Capacitor backed memory SRAM got no power due to shorted cap Symptom: Display malfunctions Cause: One display board was cracked, another had a bad connector and the last one had a bad 4511. Symptom: "Gameplay" and scoring is erratic Cause: MANY playfield switches were misaligned and thus closed when they should not have been, confusing the hell out of the CPU Big Town (Playmatic, MPU 100% identical with Last Lap) Symptom: Self test hangs at 5 Cause: Two playfield plugs were swapped Symptom: No timing (hangs after self test completes) Cause: Power supply one phase had no contact. One should assume that with half the power, the MPU should work unreliably if at all, but it worked just fine. Symptom: Gameplay erratic. Won't keep scores when turned off even with new cap in there. Resetting highscore goes to 500000 but says 502000 after short time Cause: Probably the RAM... that's gonna suck, these are hyper rare. Pinball Champ 82 (Zaccaria) Symptom: No coil power Cause: Dry joint on one of the big rectifiers. Anyone could have fixed that, but since I do rare exotic pinball machines... Middle Earth (Atari) (not LOTR related) Symptom: Just dead. GI turns on, that's all. CPU board appears dead under the scope, too. The quartz, the LS00 and the LS93 work. Cause: 9602 CPU Clock pulse shaper dead. NOTE: The 9602 is unobtainium as is information on this chip other than dual monostable multivibrator. This smells like an LS123 with a different pinout, so I made an adapter and guess what - IT WORKS! That is if you pull Input A to LOW instead of HIGH. Symptom: Still no clock on the CPU Cause: LS74 after the pulse shaper has !Q stuck high Cause: It wasn't the LS74, it was the LS14 that it had to drive, its inputs (many of them) were high impedance. Break: Just out of curiosity I checked all ICs on the board that I could with my fancy new old In Circuit Chip Tester/Debugger and it found another bad LS14. (it had only one of 6 gates bad, that was the Sound Enable line) Symptom: Reset goes high for a second, then gets stuck low Cause: My version of Middle Earth has the Power Fail transistor which gets its OK from an external board so on the bench, this transistor blocks the CPU. Also, without it it would be permanently watchdogging. Bench Cure: Bent pin 6 out from that other LS14 that also does Sound Enable and put Ground into the socket. Symptom: Now I have a random assortment of endless loops. Cause: Bad RAM and bad ROM. My chip tester told me that the first 2111 was dead as a doorknob and the second was horribly senile. Made an adapter for 2114 RAM/ Symptom: Still garbagey endless loops, but counting stuff up. Cause: Bad ROM. Dumped the ROMs, sadly my revision was different, but one of the ROMs had two address lines stuck and putting in the replacement ROMs of a different revision brought the machine back to life. Symptom: Crashes after a few hours of playtime, reboot only fixes it for a few minutes, after a couple of retries the machine just stays dead Cause: The already socketed LS74 that had to fight the bad LS14 was the culprit. Machine is now rock solid. Symptom: Most solenoids stopped working, some strings of lights don't turn on Cause: Bad edge connectors Symptom (after 3 years of working): Gives 5x scores even though 1x selected Symptom: Can't enter test mode Cause: 9301 generating keyboard strobes for the dipswitch bank had one output dead. Replaced it with a freeform socket and an LS42. Symptom: Completely dead. Sometimes runs a short endless loop. A2 stuck low. Cause: Bad 74367 BUS driver. Pinball Lizard Sound board Symptom: After game starts, no sound effects, just an endless loop of thunder. Cause: The LLU (tongue/lamp) board had a few chips gone bad, especially U10 which I think was a 324 Op Amp. Symptom: Sound effects work, but only for an hour or two, then it's back to thunder loop Cause: Bad tantalum cap on the LS121 (forgot which U) Spiderman (Gottlieb) Symptom: No boot Cause: Not found, works on bench, probably contact related Symptom: Garbage on displays Cause: One of the three LS48 was bad. Symptom: No/erratic sound Cause: Bad IC sockets, bad contacts. Lost World (Bally) - they keep making movies out of old pinball machines! Symptom: No boot Cause: There was another 6800 CPU where the second PIA should have been. Sadly, this little free extra CPU was dead... Symptom: Optically, no boot (but would work though) Cause: Driver transistor for LED falling apart. Symptom: Contacts erratic Cause: Old machine hasn't been serviced in a few decades. Bad contacts. Symptom: Dragon's Den contact comes through as closed even though it's open Cause: Switch capacitor bad even though it passes the tester, also another switch was at a time replaced with something differend and had the diode wired in a nonstandard way. Haunted House (Gottlieb System 80) Symptom: Half boots but any key press yields a glitchy "24" in player 1's display. Symptom: Dead on bench Cause: Bad cap in reset circuit not causing the power on reset. Cause for switch problems: Bad RIOT. Cobra (Bell Games) (weird it worked when it was put into storage) Symptom: Signs of insanity. Display strobe is so slow it's visible, counts coins with no coinup, crashes randomly. Cause (the guy checking the machine said he probably slipped shorting the lamp matrix to the switch matrix): Blown 6821 PIA. Unknown Bell Games CPU board Symptom: Dead Cause: Bad battery (though relatively new) drags down Vbat so low the SRAM can't work Lost in Space (SEGA) Symptom: Sporadic random resets for the last few years, now it resets every few seconds. Cause: Zero Ohm resistor shorted to a piece of wire patching a trace underneath it that was eaten away by a leaky battery Blackout (Williams) Symptom: Pretty much alive biologically speaking Cause: Stored in a VERY moist place... this is gonna be a nightmare to fix playfieldwise. Visual inspection and testing all socketed ICs (all the driverboard ics were socketed) shows 2 defective 74LS... on the driver board, one pre-driver transistor missing and one driver transistor just glued on, as well as a burnt up trace. Symptom: No displays, no attract mode, self diagnosis flashes once, then bottom LED stays on Cause: Socketing and testing all RAM chips revealed the 5101 is bad. Symptom: Testing driver board in known good machine, plays sounds in endless loops, swapping CIAs around blew the coil fuse Cause: One of the CIAs bad - test machine works perfectly now. Symptom: Master display defective Cause: UDN6184 for column strobes defective Symptom: Player 1 display defective Cause: Bad plasma tube. Symptom: Sound board: No sound. Cause: All ROM sockets bad Symptom: No sound Cause: Sound OS ROM comes back empty (it's a mask ROM...) (sound works now) Symptom: Two controlled coils don't work Cause: Maybe bent wires shorted the inputs to the pre-stage, as I bent them over to investigate, everything started working Symptom: One row of lights not working Cause: Pre-stage driver transistor had a broken leg. Symptom: Some switches not working, some switches making permanent contact Cause: Long storage in a moist place, I would say. The machine plays OK-ish now. Actual Cause: Switch matrix strobe bit 0 made no contact with the CIA driving it (bad socket) so was stuck low. Symptom: Machine goes haywire if played cold Cause: I'd guess the original capacitors aren't that great anymore. Symptom: Sudden Death (braindead) Cause: LS139 chip select generator. Symptom: Sometimes weird scoring Cause: Switch matrix is missing one strobe Cause: Bad PIA socket Blackout (another one) Symptom: middle pop bumper stuck on Cause: IC6 (LS08) bad Symptom: Error code for 5101 bad Cause: no idea. I swapped RAM with another machine and both pass self test Symptom: Noisy sound Cause: Bad blocking capacitor next to the 7805 Symptom: Middle pop bumper doesn't work Cause: Previous owner completely changed the circuitry after the LS08 failed Symptom: Upper 16 strobes for the displays intermittent Cause: Bad solder joints, marginal UDNs World Cup (Williams 1978) Symptom: No boot/crashes/intermittent boot Cause: Bad chip sockets Symptom: Not booting or very flaky Cause: Lots of dry joints on power supply module Symptom: Right slingshot locks on Cause: Bad 7408 Symptom: MPU not working/erroring out Cause: Someone (not me) put a 6810 RAM in backwards (it survived) Space Race (Recel, mechanical pinball machine) Symptom: Not counting down bonus, tries in an endless loop Cause: Bonus relay one of the switches broke off Take Five (Allied Leisure 1977) Symptom: Doesn't eject the ball Cause: Coil bad, driver transistor bad, pre-driver transistor bad Symptom: Score displays only show 000000 even though game is running Cause: Neither the clock nor the data line to the display were working Cause: Bad LS42. Symptom: Player 2 display only the ten's digit works. Cause: First LS164 shift register on display board Symptom: No knocker Cause: Blown resistor and bad pre-driver transistor Cosmic Flash (NSM) Symptom: Died, LS06 next to CPU desoldered itself, got super hot Cause: Unknown. Replacing the chip made the machine work for a few hours Symptom: Died again (this is when I got the board - LS06 replacement done by owner) Cause: Several of the traces/vias on the LS06 were bad. Owner says machine is running fine now Tri Zone (Williams) One of the driver transistors failed the Testofon test. Symptom: Dead. Flashes test LEDs, but dies immediately. Cause: SCANBE IC sockets. Cause 2: The two 6889 next to the CPU were said to have gotten extremely hot. Their outputs look a bit slurry, but match the inputs. Replacing the ICs with wire bridges as seen in my Blackout made the CPU board work again - with Blackout ROMs Symptom: Not working with original ROMs Cause: GAME ROM bad Symptom: Sound board starts buzzing after half an hour Cause: Big 1200æF electrolytic in sound board leaky Symptom: Only goes to Audit menu, even with battery installed, doesn't react to button presses Cause: Both 4049 in the keyboard matrix bad Symptom: One string of lamps way too bright Cause: 27 Ohm resistor was replaced with 10 Ohm resistor Symptom: Still way too bright Cause: Bad PIA Symptom: Dead Cause: A7 to the CPU has no contact, had to replace CPU socket again Data East Star Wars Symptom: "I downloaded a hacked ROM, now there's a feature missing and then it started smoking and the lower playfield GI turned off" Cause: Bad TIP42 in light strobe, must've shorted to solenoid voltage The pre-driver survived so there's a good chance the PIA is OK Symptom: Lights appear to work now, starts a game, but no flippers Cause: Relais driver transistor shot Cybernaut Symptom: You have to turn it on for 15 minutes, then off and on again, then it works. Cause: Gremlins. Never found. The LED turns off after first power up. Replacing the bad ROM sockets solved the problem exactly once. Now the machine is back with its owner and I have no chance at trying it again. Robot (Zaccaria) Symptom: Dead, knocker coil locked on Cause: Bad LS00 in the CPU address decoding Symptom: Keeps blowing the 39V fuse Cause: Probably knocker coil shorted, knocker coil transistor was dead short. Symptom: Some lights not working Cause: Bad ribbon cable, missing chip select for one of the DEMUX ICs.