Saturday 27 February 2016

Crazy Kong repair

This board had two faults, various graphics glitches and noisy / incorrect sounds playing. I started with the graphics fault first as thought this would be the harder to find.

At the start of the game Kong climbs a ladder, the sprite was messed up, Mario looked weird when moving like he had a broken leg and when he got to a certain point on a ladder he turned into a blue ladder.



After ruling out the usual suspects (bad traces, eproms and fujitsu ttl) I was left looking at all the other ttl, and there's a lot of it. The Gauck comparator probably would have saved me an hour or so but unfortunately I haven't even started to build it yet. 

Since the fault isn't constant it was tricky to track down, I decided to crash the game at a point where it was glitched, I did this by bridging two address lines on the Z80. Then went around the board with the logic probe and hp comparator, I eventually found a 74LS273 @ D2 with output pin 16 stuck low.

It's not a Fujitsu for once.



I replaced the 74LS273, Kong is now restored and Mario can run and climb ladders without disappearing once again.

This just left the sound issue which was a simple one, I replaced the AY-3-8910 chip and all sound was restored and played correctly.

Wednesday 24 February 2016

Bank Panic repair

I've had this board for years, I got it when I cleared out a local operator. I always thought it was a puzzle game for some reason but after watching game play of it on youtube I finally decided to dig it out and make yet another jamma adaptor.

The label maker ran out of tape. :(


When powered up I was greeted with a black screen and no sound. Checking the cpu (Z80) with a logic probe I could see the reset circuit was working as it should but there was no clock, it was stuck high instead of pulsing.

Looking at the crystal it looked fine, and not snapped off as I often find them. Looking slightly above the crystal I see a Fujitsu 74LS74.

The comparator flagged pins 5 and 6 (outputs) as bad.

Piggybacked a good 74LS74 on top.

and it fired right up.

Unfortunately after about five minutes of playing the game, it crashed and reset then refused to work at all.

Looking over the board again I spot more Fujitsu.

 Another 74LS74 and again it flags the same two outputs.

I was going to replace all the Fujitsu anyway, it's like they knew and committed suicide as I found two more bad 74LS74s.

All Fujitsu removed and replaced.

The board is back to working.

The sound was a bit scratchy and quiet but after replacing the 3x 220uf 16v caps in the bottom right corner of the board the repair was complete.

The game is surprisingly fun and addictive, defiantly one for the collection.

Friday 19 February 2016

Donkey Kong Jr (Good working order)

I just picked up this Donkey Kong Jr board from ebay, described as good working order. It certainly wasn't by the time it made it to my door step, it didn't help that it wasn't even put in a box. Bubble wrap and brown paper... typical ebay.

I didn't bid much on this as I thought there was a good chance it probably didn't work given that his Nemesis board was listed as untested / unable to test.

I did think it was an orignal Nintendo board though as it looked the same as an original from the low res pictures and I didn't know bootleggers did 1:1 copy boards. Turns out they did, all they changed was adding an audio amplifier and out putting standard video instead of inverted. This is better in some ways but it also means I cannot use my Mike's Arcade Nintendo to Jamma adapter.

First thing I had to do was wire up an adapter.


I've had to made a lot of these over the years and it never gets any less tedious.

Before hooking it up I had a good look over the board, I noticed a pad was missing on the edge connector, looking at the pinout this is Green. So it is going to need to be sorted, I lifted a pad off of a scrap board and glued it down then ran a small piece of kynar wire. 
It's not winning any beauty contests but it'll work.

I power it up and got no sync, this is a common problem on original board too. I knew this already from a previous repair, the pots on the bottom board get damaged and the board won't sync. This can happen when you post it without putting it in a box first!.

Resoldered the two legs that had lifted on the vertical pot.

 Things aren't looking good

or working.

Now I decided to reseat all the EPROMS and check them with romident while I was at it.

EPROMs 7E and 7F both had data romident didn't recognize.

These EPROMS didn't look original to the board, looks like they were just thrown in to complete it. Testing again sorting these two roms didn't help much.


After looking at the 2114 rams everything looked ok so I decided to go Fujitsu spotting. These two 74LS74s on the top board were the first two I looked at.


I stuck the HP Comparator on the top chip (7F) as it was right next to me and even had the 7474 card in it.

It lit up like a christmas tree.


I removed and replaced both 7474s as they were together and both Fujitsu but only 7F was bad.

It looks like MacGyver has worked on this board at some point, what do you do when you lose the pcb spacers?



 Melt a bic pen and hold the pieces in place with wire of course.

Replaced with some cheap spacers from China.

 There was also this strange ram bodge.

He looks so lonely.


 Game is now in good working order.

Wednesday 17 February 2016

Desoldering and Fujitsu TTL

I desoldered all those nasty sockets from the Hyper Olympic / Track & Field board mentioned earlier.


Which unfortunately made no difference to the operation of the board, but it needed doing. I have a feeling I have at least one bad faulty custom chip but the board is littered with Fujitsu logic so maybe not all hope is lost for saving this one.

Speaking of Fujitsu logic while I had the desoldering station fired up I decided to remove all the evil F chips from an Aliens board that just constantly resets, to see how many were bad.


Out of the 29 Fujitsu chips 'only' three of them were bad (the three on the far right). I've got a couple of TMNT boards that will get the same treatment eventually but I need a break from desoldering after this lot.

Sunday 14 February 2016

Kung Fu Master PCB Repair

I bought this non working original Irem Kung Fu Master pcb from griffo on ukvac forums. I've always liked this game since playing the NES version loads back in the day, I actually like the sounds better on the NES but obviously the graphics are a lot better in the arcade version.

This is what was displayed when hooked up.

It's Z80 based board so time to hook up the fluke.

The fluke failed to test the ROM and RAM.

So I had a look at the logic between the CPU and ROM/RAM.

Four Fujitsu 74LS367s

Fujitsu TTL is well known to fail far more than other manufacturers for what ever reason. I decided to remove all of them now as even the working ones will fail sooner rather than later. I also replaced a Fujitsu 368 next to the CPU which just leaves three more F chips on the board which I will order and replace at a later date.


Thomas is back to saving Silvia.

Friday 5 February 2016

Track & Field / Hyper Olympic Test Rom

I've had a Hyper Olympic pcb in my to do pile for years and keep putting it off as all the sockets on the main pcb need replacing and all but three of the eproms were missing. After seeing shoestrings test rom posted on klov I decided to give it a try. First I had to replace all the missing 2764 eproms and then burn and fit the test rom.

The latest version at the time of writing this is TEST 1.1 and isn't in the klov post but is on jammarcade.net here.

Picture doesn't show up too well but a screen came up with 11A NG!

Looking at the board 11A is a 2114 RAM.

Desoldered it and sure enough it was bad.

I fitted a socket and new 2114 powered it on again.

The test rom now runs with sound. NVRAM shows up as bad as the battery has been removed.

Powering on with the original rom in 5A and it still doesn't run, which wasn't really surprising since there as so many bad sockets on the main board.

I don't have the sockets or the time to finish this at the moment but I'll update when I get back to it.