20 Jun 2018

Space Invaders Deluxe - Midway 1979 (repair log)

This board was sent to me in order to convert it to Space Invaders. Owner said it should work...

Upon reception I quickly noticed 2 things:
- all ROMs were pulled
- one chip was missing on the sound board

By looking at pics online I found the missing chip was a NE556 however I had only NE555 in my stock so installed a socket and waited for the part to come in the mail.


Then I strapped to board to use a single 8kB chip for the program ROM (really quick and easy to do and saves you the hassle of rotten sockets) and installed a DIP28 socket for ROM H instead of the DIP24 originally installed. I also merged the 4 Space Invaders files together and burnt the resulting file in a 27C64 EPROM, installed it in the socket and powered the board: game was stuck on a garbage screen with lot of flickering dots. For sure there was something wrong with RAMs.

I pulled the EPROM and looked for the solid lines pattern those boards should display when no program is executed: only garbage and flickering dots. Probing the RAMs revealed at least 10 (out of 16) were bad. I simply pulled them all and ordered 16 new ones for the peace of mind.


The next week the fresh RAMs showed in the mail: I installed DIP22 sockets (best move ever) and saw a lot of improvement in the pattern screen with some solid lines but also still a lot of incompletes/flickering lines. I probed the RAMs again and found 7 were bad. Upon inspection I discovered those NOS RAMs ware more likely pulled parts... Unfortunately it was too late to make a claim (I ordered them months before getting back to the repair). I ordered some more from a different place (did I tell you they are becoming crazy expensive because of rarity these days?). Rince and repeat: 10 chips received, 5 bad... Ordered again from a third seller, much more expensive than the 2 previous ones. This time out of 6 ordered chips I received 6 working ones (you get what you pay for...).
This time I finally got 16 good RAMs installed on the board (and one spare):



Success!

I installed the missing NE556, tested sound effects and controls, and found no further issue.
Game fixed!

13 Jun 2018

Neo Geo AES/MVS battery free 16 in 1 memory card - part 1

Yes it's been done by several people before.
However every time I've tried to contact one of them to buy one of those memcard they were never available or had no answer at all...


All the needed information can be found there:
https://wiki.neogeodev.org/index.php?title=Memory_card


I simply routed a very simple PCB (FM18W08 FeRAM and one switch to select between two 32k banks, each being equivalent to 8 original memcards regarding capacity):




Now waiting for the bare PCBs to come in the mail.

6 Jun 2018

Conversions on Jaleco Mega System 32 - part 2

A collector (let me know if you want to be named here) came to my rescue... Again!
I sent him the files to convert Tetris Plus, originally using a ss92046_01 security chip, on ss92047_01 security chip. As he already owned an original Tetris Plus romboard it was only the matter of swapping all ROMs except the 2 used for background and texts graphics.

After few days of remote troubleshooting we finally had success:


Here you can see the burnt .10 and .30 EPROMs:


I've updated my script and now any game can be converted to work with any security chip by a simple click on a button.

Next conversions requiring testing are on Jaleco Mega System 1 (isn't it logical after Mega System 32?).



30 May 2018

Terra Force - Nichibutsu 1987 (Repair log)

Game worked except background plane was mostly missing:


My eyes quickly spotted "shiny" RAMs on the top board so I started to probe them:


And I quickly found video RAM @ F17 was faulty. I installed a new one and problem vanished:


Game fixed.


23 May 2018

Sega 834-6510 I/O board reproduction - part 1

As far as I know this board is used only on System 24 for driving games (Hot Rod and Rough Racer) and golf games (Jumbo Ozaki Masters and Dynamic Country Club).
It handles analog inputs (e.g. steering wheels, gas and brake pedals, etc.).


I've been looking for one for a very long time and finally took the decision to reproduce it. Still it wasn't easy to find an original board as model for my reproduction.
Once again a German collector came to my rescue and lent me his board:



And here's my CAD reproduction:


It's quite expensive to build due to ICs used being dropped for a long time (D4701, M5203).

16 May 2018

Xexex - Konami 1991 (repair log)

PCB was in very good condition but threw a "SOUND MEMORY BAD" error on boot up:


Of course this error isn't in the manual and even after an extensive online search I haven't found any mention of it anywhere...
Well after some probing and testing I came to the conclusion it had nothing to do with the RAM chips on the mainboard. I mistakenly thought it was caused by a bad '054539' custom chip (samples player) so bought a replacement pulled from an other game from Caius @ jammarcade.


I pulled the suspected faulty chip and cleaned the pads:


Then I aligned the replacement chip and soldered the 4 corners:


Then applied some SMD soldering paste:


And used hot air for soldering:


Only to discover it didn't change anything regarding behaviour of the board, still stuck on the same error.
So I had a look in the code and found this error is reported when the RAM present on the custom sound module is bad. Many of you know there isn't any RAM on this module but instead there's a '054321' custom chip handling, amongst other things, sound volume. I came to the conclusion the chip embed some RAM, which makes sense.
I swapped the sound module with a working one and indeed fault followed the suspected faulty one. I went even a step further and swapped the '054321' custom chips between modules: and boom we've got a culprit, the '054321' was obviously faulty.




Game fixed in a sense (although I don't have any spare '054321', at least fault is identified, it's just a matter of soldering a good chip now).

I then had an other discussion with Caius who told me he encountered similar symptoms on a Bucky'O Hare board few days later:


Damn it... I should have waited few more days before starting this repair.



9 May 2018

Super Pang - Capcom_Mitchell 1990 (repair log)

An other suicided game with additional problem: instead of the ligth blue screen the suicided game should display I was greeted by a black screen...

I followed the desuicide procedure found on the "Dead Battery Society" website (http://www.arcadecollecting.com/dead/), and then fired the board: black screen just as before but audio and controls were working. Game was playing blind, I could add credits and start a new game.
When you have no video at all it's always good to start from the JAMMA connector and then follow the signals. In this case red, green and blue signals on the edge connector were stuck low. I then probed the colours RAMs (two of them) @ 8C & 9C: all data pins were stuck low. I took the chance of piggybacking them and obtained an image. Colours were not perfect but this is something to expect when you piggyback chips as the faulty chips underneath can drag the signals low or high and the connections aren't at their best.


I pulled the faulty RAMs and both turned out to be fried on my programmer. I installed sockets and put fresh RAMs in them and was greeted by a fully working game:



Game fixed.