New Life Games Tech Forums

Model: Kangaroo
Symptom: Monitor only shows a white screen

This was an old WG 4600 series monitor. Replaced a shorted zener diode and blanking transistor on the input daughterboard. While the board was out a cap kit was installed and all cracked solder joints were fixed.

Model: Pit Fighter
Symptom: Game controls not working

Replaced bad 74LS244 ICs at locations 70B and 75B. Tested board.

Model: Pit Fighter
Symptom: Garbled graphics

Swapped IC S30 with a parts board to fix. Burned a replacement GAL for the parts board.

Model: Pit Fighter
Symptom: No video output

Traced the missing video signal back to the output latches. Signal was on the input side but not crossing the IC. The clock signal was missing on the latches. Replaced the GAL at 65D to fix.

Model: Pit Fighter
Symptom: Loses most sounds after heating up

Most audio signals disappeared after playing the game for about 10 minutes. Cooling spray had no effect. The audio signals would come & go. Sometimes they would work but most of the time they wouldn't. IRQ lines were going dead when the sounds disappeared. Replaced the 6502A CPU on the sound board to fix.

Model: Any old Atari game with the Big Blue caps in the bottom of the cabinet.
Symptoms: Locks up, reboots randomly, humming in speaker, and more.

The big caps in the bottom of the cabinet are notorious problem parts and should be replaced if you aren't sure they have already been replaced.

While checking the power supply be sure to also check the audio/regulator boards as they are bad about having cracked solder joints.

Model: Pole Position II
Symptom: runs but has garbled video

Put the game in test mode and found that the video RAM was bad. Replaced 2 2148 SRAM chips and had fun playing the game.

Model: Road Riot 4WD
Symptom: No sound

Volume was set to zero. This model is strange in that you must go into the electronic setup and choose the volume setting. There is no volume control.

If the EEPROM used to store the settings is bad then this setting will not be saved. The sympom of this is if you set the volume and other settings and they aren't retained if the board is powered off.

Model: Road Riot 4WD
Symptom: No sound and corrupt graphics

The MOB (Motion OBjects) checksum test failed. The EPROMs all had a checksum of 0000 meaning they weren't being accessed. The motion objects on the screen were missing.

Replaced bad GAL at board location 11S (part # 136089-1009) to fix graphics problem. Replaced missing GAL at location 6C (part # 136089-2011) with a programmed PALCE device (functional equivalent once programmed) in an attempt to fix sound. Sound worked but would either be at maximum volume or at zero. The settings would not save. Replaced bad 2816 EEPROM IC to fix the problem with saving the settings.

Model: Several different Atari games
Symptom: Bad GAL chips

The GAL chips on the Atari arcade games are set with their protection bits on. This precludes reading a GAL chip from a working board to fix a dead one.

Replacement code for burning your own GAL or PALCE chips can be downloaded from the Atari GALlery web site  (http://www.aarongiles.com/gallery/index.html) for the following games:

• Guardians of the 'Hood
• Hydra
• Moto Frenzy
• Pit Fighter
• Relief Pitcher
• Road Riot 4WD
• Shuuz
• Skull & Crossbones
• ThunderJaws

Model: Road Riot 4WD
Symptom: Corrupt Graphics

The title screen on the game was skewed for the top half of the upper set of words ("ROAD RIOT") and during the game the sides of the road were randomly split and moved.

The custom IC at board location J18 (Part # 137419-104) had a broken leg. Swapped out the IC from a parts board to fix.

Model: Shuuz
Symptom: Garbage on screen.

A bunch of numbers flashed on the screen repeatedly and the game would reset. It was "stuck in watchdog: with a message at the bottom saying "TOO MANY EETRIES"

Tested the EPROMs and SRAM ICs. Replaced 1 bad SRAM but was still stuck in watchdog.

The MAME .c driver showed the game would initialize the EEPROM as the first thing. Checked the 28C16 with a logic probe. The chip enable, output enable, and read lines were OK. Pulled the chip and tried reading it in my EPROM programmer. Every time it read it gave a different data set out, in other words a different checksum showed on the programmer each time.

Replaced the 28C16 and the game fired right up. Turns out "TOO MANY EETRIES" is an error message on a core dump for the game.

Model: Tetris
Symptom: Stuck in Watchdog

Game had no display. Checking the address and data lines with a logic probe revealed that the game was stuck in watchdog - it was constantly resetting.

Replaced a bad 2804 EEPROM IC and reinitialized the game's settings.

Model: Area 51
Symptom: Dead

Game board was stuck in watchdog.

Board had been worked on before and was tagged as "unrepairable". One chip that was replaced not soldered well at all. It was like they used silver solder on it and it was VERY difficult to resolder. Resoldered that FPGA chip then the CPU and DSP chips.

Cleaned the PLCC socket on the IC that reads the gun inputs and replaced a bad 14.31818 MHz crystal oscillator.

Replaced the missing hard drive with a 2Gb Compact Flash card and adapter then tested the game.

Model: Centipede
Symptom: Garbage on screen. Enemies move erratically

Replaced bad RAM at M5. Logic probe showed erratic signals on DO2 (Data Output 2)

Model: Centipede
Symptom: Garbage on screen, reboots at random

Fired up test mode and received 9 beeps. Replaced the bad RAM at N7 and tested. Garbage was gone with the exception of horizontal lines on the screen. Cleaned the character ROM legs to fix the lines. Cleaned the program ROM legs to fix the rebooting issue then tested board.

Model: Centipede
Symptom: Dead

Replaced a missing clock crystal and tested. Board came up with a white screen and CPU stuck trying to boot. Cleaned the legs on the program ROMs then the board would try to come up and had some random characters on the screen. Found where something scraped across the board, dragging some solder with it and shorting 2 traces. Removed short and tested the game. Replaced 2 broken transistors and retested.

Model: Centipede
Symptom: Board hacked

The MC14584 chips buffering the trackball inputs had been modified. Removed the 2 chips, 6 tacked on resistors, and 6 tacked on capacitors that were sticking out above the board. Removed 4 3.9k ohm resistors that were changed on the board. Installed 4 10k ohm resistors and 2 new MC14584 chips. Tested.

Model: Area 51
Symptom: Red/Purple scrambled screen

Board had no output on sync. Replaced a bad 7406 at location C20 to fix sync. A 10 ohm SMT resistor at location R118 was out. This caused the board to have no green output. Replaced the resistor and tested the board.

Model: Dig Dug
Symptom: ROM 1 Error

Checked the ROM at 6L and it had a bad checksum. Replaced ROM and tested.

Model: Centipede:
Symptom: Screen full of lines

The game powered up and the screen looked like an Atari 2600 without a cartridge installed. Checked the clock circuit and replaced bad 74LS163 ICs at M2 and P3. Board powered up but the screen was partially garbled. Replaced bad RAMs at L7, M7, and N7. Tested.

Model: Centipede
Symptom: Stuck in watchdog

Cleaned the pins on the CPU and ROMs... Replaced the sockets and cleaned the pins on the POKEY and NVRAM chips. Tested board.

Model: Centipede
Symptom: Dead. Nothing on screen.

Board was stuck in watchdog and had no output to the screen. Cleaned the legs of the CPU, POKEY, and ER2055 EAROM. Board would give a single beep when put in test mode now. Replaced the 2114 RAMs at F2 and H2. Board would then play blind. Board was missing the /HBLANK signal. Replaced bad 74LS107 at L4. Board then had video output but it was slightly garbled. Replaced the RAM at M5 then tested.

Model: Area 51/Max Force combo
Symptom: Player 1 gun not registering hits

Board had attempted repair damage.

Replaced bad surface mount 74LS14 on the gun inputs. Patched damaged traces, and tested.

Model: Centipede
Symptom: Player 2 stuck to the right side of the screen

Very odd symptom. Player 1 and Player 2 use the same controls and input circuitry. Replaced bad 74LS257 at D9.

Unsure why it was working perfect for player 1 and not player two! The chip showed bad on all 4 outputs on a BK 560A IC tester, but still worked fine for player 1. Head scratcher for sure!

Model: Centipede
Symptom: Dead. Solid white screen.

Board was dead. Data bus line D1 was stuck low on the far side of the 74LS245 latch from the CPU. Replaced 74LS373 at J5 and would play blind but the sounds were off. Replaced bad POKEY to fix audio issues. Replaced bad 74LS289 Bipolar RAM at C8 to fix white screen issue. Board would play, but the archer would not move fully to the left and centipede would get stuck on the left 1/4 of the screen. The board would not pass self tests and would continuously beep. Used a Fluke 9010A to test the board and found RAM range 0500-05FF had a stuck bit. Replaced bad 2101 RAM at K5 and tested board.

Model: Centipede
Symptom: Dead. Black Screen

Board would not read the RAM or ROMs with the Fluke 9010A. Found smeared solder short under ROM at J1. Fixed solder short and replaced ICs B1, C1, F2, H2, J2, J4, and P6 that were damaged by the solder short. Board was still dead, but the CPU was able to read and write to all of the RAMs and read the ROMs properly. Discovered board was missing VSYNC signal. Traced it back and discovered a 74LS163 at IC P4 instead of the proper 82S129 PROM. Programmed a replacement PROM and installed it. Tested board.

Model: Centipede
Symptom: Would not coin up with the correct amount

Replaced bad DIP switch and tested.

Model: Centipede
Symptom: Dead

Replaced bad ROM socket at J1. Reseated the other 3 ROMs. Tested.

Model: Centipede
Symptom: Stuck in Reset

Board would boot to the game but constantly reset. Replaced bad RAM at H2 and tested.

Model: Shuuz
Symptom: Excessive EEROM Retries error. Missing motion objects.

Replaced 28C16 chip. Replaced missing GAL chip at 97N with one programmed with code from the Atari GALery. Tested.

Model: Area 51/Max Force Combo
Symptom: Will not boot.

Cleaned soda spill from board. Replaced IDE to Compact Flash Adapter. Tested.

Model: Centipede
Symptom: Dead. No video.

Board had CPU and ROM activity but no horizontal or vertical sync or blanking signals. Replaced dead 74LS163 at N2 and tested.