What fixed your game today / general repair log

4th of July weekend, I noticed the red color was not appearing on my Lucky and Wild game. A kind KLOVer or 2 (mecha and tomservo) said it was probably a cold solder joint on the monitor chassis.

Today I finally had a chance to pull the chassis and check it out. Sho nuff, the solder on the underside for the red pin (outermost pin facing the back wall of the game) had like 0 solder on it.

Flowed new solder and now the red is back as it should be. I've owned this game for ~17 years. Thanks to you guys, we should be able to make another 17!
 
fb019f13-943a-4a01-adf4-f7e405ce5ba8.jpg

Xevious at work a few nights ago froze up on the scrambled gibberish boot up sequence. by time I got out to the floor however the game was running again. I proceeded to go through the whole battery of pulling the boards and pressing all the chips back in. not much crunch. the board edges needed cleaning, I formulated a new procedure for cleaning those contacts by using Novus #2 (which is a fine product I normally use on pinball) and the amount of black gunk you get on the towel after is incredible. I do a follow up cleaning with rubbing alcohol. after some voltage adjustments, the game works again.

until I came in Friday night, it's on the broken game sheet as saying it's a black screen. pulling the game out and turning it on I definitely heard the monitor running, so it wasn't that. I had no LEDs on the game CPU/video boards. the 20A fuse was blown. now it's a murder mystery. I pulled the 9 pin plug from the AR2 and replaced the fuse and turn the game back on and had life on +5V. I escalated to connecting the 9 pin plug to get the game back in the mix. the board LEDs didn't immediately turn on, there was a delay, never seen that before. but the game is running. I proceed to play it for awhile and then circle around to the back again. now I smell burning electronics aroma. yes... that's the bridge rectifier under the power brick.

524412037_1252168356071138_4851926499665055050_n.jpg

you can see the quick disconnects are all barbecued. this is entirely normal and should be something to check on all Atari games. these have insulated quick disconnects but Atari apparently used uninsulated ones also. after extracting the bridge rectifier I started with black lead on the + side and red lead to the AC sides. junction drop on one, short on the other. the bridge rectifier was just bad. one of the age old questions I've asked on my journey here for a great many years was how guys just wrapped fuses in foil and kept running games with shorted bridge rectifiers. what I learned is the game can apparently still work under such conditions. I guess props to Atari for their power design, the 20A fuse did its job in preventing mayhem.

as we don't have 5010 bridge rectifiers but several power bricks on the shelf in the back I was able to find a donor that tested good. I'd intended on soldering the wires to the donor bridge rectifier, I had enough length for the AC tabs, but the + and - from the filter capacitor were not long enough. so I wound up robbing those wires from the dirty parts brick also.

in case anyone was wondering how this affects the +5V output, with a healthy bridge rectifier it came out to about .10V less than with the shorted one. I believe I had 4.99V to the video board (I measured at the axial capacitor) and that's the way I left it.

alas, Xevious was saved. I hope this was educational cause it certainly was for me to finally put my electronics ninja skills into practice for a change.
 
Did a quick one before bed because it's been on my bench for too long.

Game pcb: PC10 Metroid Cart
Symptoms: does not appear in menu, works with a modded bios
Troubleshooting steps: There are two things that a PC10 cart needs in order to be recognized by the machine - the name ROM and RP5H01 security ROM. I don't have an easy way to read RP5H01 so I pulled the name ROM at U5. Putting it in my T48 it complained about a bad pin 28. After cleaning the legs of the eprom it still complained. Reading it anyway I got back all 'FF'. I burned a new 2764 to replace the Name ROM. This resolved the issue.
Screenshot_20250805-222301.png
Screenshot_20250805-222242.png

Solution: Bad name ROM at U5.
 
Jungle King:

Symptoms, music works, but no sound effects.

Tested all the AY8910 sound generators in a known working boardset, all work.
RAM associated with the sound generators (2114s at ic61 and ic62) test good in the retro chip tester
Pulled the Z80 at ic52 and tried another known working one... all sounds and music are restored.
 
Centipede : Garbage/Squiggles on screen.

Started with the clock... found 3 163s bad at M2, N2, P2.
Replacing those got SOMETHING on the screen that was recognizable... but there was a lot of repeating characters running down the screen...
RAM test said N5 was bad.
Replacing that got the screen cleaned up, but the player's character, shots, and all bugs were flickering VERY badly and were stuck at the bottom of the screen.
Found ANOTHER 163 @ A5 that was bad.
After replacing that the board runs well.

... and since NOTHING is ever good enough, and I never learn... I went to clean up the socketed chips... lost 2 game EPROMs and the EAROM. I had a replacement EAROM and I had a single ROM Centipede board that I made a couple of years ago and never tested. Well, it works... so that's nice.

She's back up and happy.
 
Last edited:
Fixed the 6100 in @CarrieZ 's Tempest today. Monitor had a very unstable image on powerup that would eventually become stable after 10-15 minutes. (The image would jump/expand and contract on both X and Y sporadically). Figuring it was an HV issue, all signs pointed to a cold solder joint and/or connection issue.

Fix turned out to be a couple things. One of the wires that attaches to the focus block was pretty brittle and came off when I removed the protective rubber cover. The amp connector running from the HV cage to the neckboard, had a burn mark and the pins weren't super snug so I crimped new ones. The red wire running to the neckboard was only soldered on by a strand, so I cut & stripped a new piece and resoldered that back on. Did the same with a couple of the other wires. Reflowed a bunch of the HV board and most of the neckboard. Picture is now rock solid.
 
My Arena pin had a blown fuse at F2 (power line) after problems removing the original battery.

Hoping for good vibes tonight with a new fuse. Deets in repair section. Stay tuned.
 
Every single machine other than this one hated me today. But I got a Data East Star Wars pin working.

Game: Data East Star Wars Pin
Symptoms: not booting
Troubleshooting step: power supplies on these data east machines ALWAYS have problems. Usual suspects were found, broken fuse clips, blown out caps, missing 5V. Replacing the caps and fuse clips brough me back my 5V. Next peoblem, board was still not booting. Oops, battery damage on the mpu board because it still had it's original batteries. After cleaning and neutralizing the battery damage, I noticed that it damaged the rom socket. After replacing the rom socket it finally booted.

Now I missing several coils. Hey guess what, broken fuse clips on the PPB driver board. Who would have guessed? Replacing the fuse clips brought back the coils but the VUK coil was constantly firing. That ended up being a bad TIP36C at Q3. Very common failure. With that, most of the pin was now working.

Solution: fuse clips, replaced bad caps, replaced bad Q3, replaced bad rom socket.
 
Every single machine other than this one hated me today. But I got a Data East Star Wars pin working.

Game: Data East Star Wars Pin
Symptoms: not booting
Troubleshooting step: power supplies on these data east machines ALWAYS have problems. Usual suspects were found, broken fuse clips, blown out caps, missing 5V. Replacing the caps and fuse clips brough me back my 5V. Next peoblem, board was still not booting. Oops, battery damage on the mpu board because it still had it's original batteries. After cleaning and neutralizing the battery damage, I noticed that it damaged the rom socket. After replacing the rom socket it finally booted.

Now I missing several coils. Hey guess what, broken fuse clips on the PPB driver board. Who would have guessed? Replacing the fuse clips brought back the coils but the VUK coil was constantly firing. That ended up being a bad TIP36C at Q3. Very common failure. With that, most of the pin was now working.

Solution: fuse clips, replaced bad caps, replaced bad Q3, replaced bad rom socket.
Awesome, may need to recruit you when my Arena board comes back.
 
Game: Centuri Pheonix
Symptoms: missing music
Troubleshooting step: Music is generated by MN6221AA melody generator. I made a wild guess by swapping out the melody chip because I had a non-working pheonix board. That fixed it.
Solution: Replaced bad MN6221AA melody chip
Thank you for sharing that!
 
@CarrieZ and I recently discovered that her Tempest was producing very low volume. We really don't know when the issue began because for months (maybe years) we've been running Scott's Tempest multi FPGA and have had adequate volume/sound. Recently, we were trying Jason's multi FPGA and the audio was so low that we initially thought it was a problem with his FPGA. So we threw her original PCB in and realized that the volume was also very low and even with the cabinet volume POT at max, it was too low. On top of that, I noticed that the sounds were slightly off as well - almost lowfi and kind of distorted.

Wiring appeared good, and the POT checked out with full range (1.5-50ohm). So, Carrie pulled the ARII from her Gravitar cabinet for testing. Swapped it in and we had perfect sound again with complete adjustable volume.

Solution:
One of the two TDA2002 amps seemed suspect after testing/comparing with a DMM. Swapped both out with a couple TDA2003As and WA-LA. Back in business.
 
Game: Nintendo Vs Board (mystery ROMs)
Problem: Solid color screen only.
Solution:
ROMs were blanks. Sockets were really bad on the bottom. Installed Excite Bike on the bottom and Castlevania on the top.

 
@CarrieZ and I recently discovered that her Tempest was producing very low volume. We really don't know when the issue began because for months (maybe years) we've been running Scott's Tempest multi FPGA and have had adequate volume/sound. Recently, we were trying Jason's multi FPGA and the audio was so low that we initially thought it was a problem with his FPGA. So we threw her original PCB in and realized that the volume was also very low and even with the cabinet volume POT at max, it was too low. On top of that, I noticed that the sounds were slightly off as well - almost lowfi and kind of distorted.

Wiring appeared good, and the POT checked out with full range (1.5-50ohm). So, Carrie pulled the ARII from her Gravitar cabinet for testing. Swapped it in and we had perfect sound again with complete adjustable volume.

Solution:
One of the two TDA2002 amps seemed suspect after testing/comparing with a DMM. Swapped both out with a couple TDA2003As and WA-LA. Back in business.
There is more story to add to this..
Carrie contacted me thinking my tempest multi had an audio issue and sent me her PCB to test or repair. I opened the box to see Scott's PCB. It turns out she or someone accidentally put the wrong one in the box. Eventually when received my multi tempest we found it worked fine. So I returned both multi boards with some tda2003s in hopes that they would fix the ar2 .
 
impressed that you recognized it was not the board they were referring too. either you have thought far enough ahead to preent this from happening or you have been stuck with someone trying to get a free repair or two so that you have a verification system.
 
LOL :LOL: Scott's FPGA and Jason's FPGA look COMPLETELY different. @CarrieZ just made a simple mistake while racing to ship out the PCB/FPGA before she flew out to MN. (One was in the cabinet, one was outside, wrapped in bubble wrap.
 
Last edited:
impressed that you recognized it was not the board they were referring too. either you have thought far enough ahead to preent this from happening or you have been stuck with someone trying to get a free repair or two so that you have a verification system.
I like how quickly this morphed into me trying to trick Jason into repairing a board that's not broken! :unsure: :ROFLMAO:
 
Back
Top Bottom