Hyper Olmpic (Track and Field) PCB repair

 Arcade Info  Comments Off on Hyper Olmpic (Track and Field) PCB repair
Dec 082020

I’ve had this Hyper Olympic board sat in my repair pile for more than 5 years.  I kept coming back to it, but never really made any progress, mainly because the schematics are poor.

On initial power up I got jumbled sprites for the runners and lines through the background, text and logos

I took a video back in 2015 here:

It was originally labelled a Hyper Olympic board, so I thought this might be why the logo was messed up so I changed it from a Track and Field to a Hyper Olympic, but that wasn’t the case and the issue persisted.

I also used ShoeStrings test rom at the time to see if it could identify any issues and unfortunately this did not help. Here is a video:

I used my logic probe back in 2015 and looked through the board and noticed some stuck pins on two of the eproms between 11C-15C, and traced this back to a faulty LS04 @ B6.  I swapped this out and the runners were back.  The only issues that persisted was the lines and faulty logo. I took a video here

Fast forward to the end of November 2020, and I picked this up again.  The background and text were missing details and the logo had wrong colours and there was jailbars as in 2016.  I was fairly confident the problem was around the roms D10-D12 @ H14-16 and the 273 @ 14G was the next part in the chain.  I took at look at that but found no issues.

If I removed D10-12 and the two rams @ 11-12H I could still see lines on the screen, so I came to the conclusion that they were not affecting the lines.

I had a look on the scope and again didn’t see anything obvious.  I spoke with a fellow collector and he suggested that I started to short pins to see if I could narrow the area a bit.  This is where I got my break through.  I started to ground the data pins on the D10-D12 roms and found that if I grounded the D0 pin on the roms I could replicate the issue.  This narrowed the problem potentially down to 2 chips:

My money was on the LS273 because it was a Fujitsu, so it was the first one I changed. Sadly the issue persisted. So I then decided to also change the LS157 @ 15G as I was confident the problem was definately here. I am glad I did because I powered on the board and was glad to see the ongoing 5+ year long board was now fixed!

Playchoice-10 repair #2

 Arcade Info  Comments Off on Playchoice-10 repair #2
Jun 172020

Another challenging PlayChoice-10 PCB repair!

The symptoms were that I could see the menu but there was no game – just a black screen. Triggering the game to start allowed the game to play blind, so pretty much the opposite of the board I had just fixed. Initially I thought it might be a sync issue.

I could see issues on the 4066 @3C and LM324 @3E but couldn’t pinpoint anything exactly.

I looked through all the logic from the CPU/PPU up to the game card, and I couldn’t see any issues.

What’s odd is that the menu system goes through the same sync chips, so you would expect both screens to be not showing, so that was a bit of a mystery to me unless I got it wrong?

I got the scope out and sync looked ok, but the colours were not. The failure seemed to be occuring just after 2E in the colour section. At this stage I took a guess and replaced the LM324 @ 3E as that tied colours together.

This did not fix the issue, but it helped me determine that there was no 12 volts coming into the colour circuit. What was baffling was that this board does not use 12 volts, but instead it uses a DC-DC converter to convert 5 volts into 12 volts and this was missing as I had a reading of 0.8v! What also made it hard was that the schematics do not cover this version of board (PCH1-04-CPU) and this specific area, so I spent quite a while working out what was connected to what.

The RK13 diode @ D5 was ok, which I also suspected as being faulty.

I replaced the IR3M03A DC-DC converter with an MC34063AP1 and still no 12 volts!

I then noticed that R32 was burning up so wasn’t sure if there was a short somewhere, but as soon as I removed this resistor and reflowed some components then the 12 volts was back…so having looked at the schematic, something along it’s path was causing it to overheat. If I again replaced the resistor it would melt.

Then I noticed that the PCB had quite a bit of corrosion on it in the colour section. I thought this was likely causing the issues so I ended up replacing:

Variable resistors VR1-3

Transistors Q12,Q13, Q15,Q16, Q18,Q19

Capacitor @ C20

These were A933, C1740 transistors, and some 500ohm pots, and a 470uf axial capacitor.

I powered up the board and was glad to now see that the resistor R32 was no longer burning hot and melting.

However the game still did not play! It just froze on the menu.

I spent an hour or so looking at issues, and in the end I tried the same game cartridge on another PCB, and found that the game cartridge had now failed! When I put a known working cartridge on the corroded PCB it confirmed that the game was working, everything was fixed and the replacement parts did he job!

The issue with my failed game cartridge (Golf) turned out to be a failed Eprom @ U3. I replaced this and the game no longer froze.

Playchoice-10 repair #1

 Arcade Info  Comments Off on Playchoice-10 repair #1
Jun 172020

A fellow collector asked me to take a look at a couple of Playchoice-10 game boards that he was having problems with, as it was beyond his capabilities to repair them, and it was his favourite machine that he could no longer play. I’ve not looked at these boards before, so thought it was a chance to get a bit more familiar with them.

On unwrapping and testing the first PCB I could see that the game played ok, but the menu was all scrambled:

I knew the Playchoice-10 BIOS, which was essentially the menu, was the rom located at 8T, so I suspected that the fault would be around this area. Using a logic probe I found a stuck pin on the ram @ 8R pin 21 in literally 5-10 minutes of looking. I traced this fault back to the LS157 @ 7V. The output of this chip is connected to pin 21 of the ram. The inputs to this chip are pins 5 and 6. Pin 5 was pulsing, 6 was high, 7 was stuck high, so pin 7 had a fault. Piggybacking a new LS157 over this chip, sorted the menu problem so it was time to replace the chip.

Pleased with my repair I put this board aside, and tested it again a few days later before sending it back to its owner. I was very disappointed to find that the board had failed again and this time it was more catastrophic! The board was completely dead! No menu and no game on either screen. The LS157 I had just replaced had also failed! Argh!

Replacing the failed LS157 brought back a scrambled menu, but the game was no longer playing at all. I literally spent hours and hours looking over this board and finally found a problem with the LS00 @ 6U that was causing all manner of issues with the read and write pins in/out of the z80 CPU. Swapped this out and the game is alive again! Hopefully a bit longer this time!

Q*bert knocker repair

 Arcade Info  Comments Off on Q*bert knocker repair
May 162020

For those that don’t know what I’m talking about, a knocker is basically a bit of a gimmick that was installed into a Q*bert cabinet. The idea is that when Q*bert falls to his death off the pyramid, the game gives out a thump as he hits the bottom of the arcade cabinet using a coil to fire the knocker. Clever huh?

Since I’ve owned this cabinet I have never been able to get the knocker working, so this weekend I decided that this was going to change.

I had previously purchased a replacement known working knocker, as the plunger that makes the knocking noise, on my original knocker, had totally seized up. When I got the replacement I hoped that it would be a case of just swapping the part and enabling dip switch 6 on the main PCB and that was it. Sadly I was very wrong!

So the first thing I did was grab a copy of the manual and have a look a the wiring diagrams for the knocker area. I also confirmed that the plunger was freely moving, which it was, and that the 1 amp fuse at F6 had not blown, which it hadn’t

Next I started to look at what voltage was required to power the knocker, and started to trace it all the way to the PCB from the Power supply. I attached my black multimeter wire to the ground strap and then I could see +30v DC at the soundboard (A6, J1-P1), and the sound worked ok, and from there went backwards through all these locations:

  • Video Power supply pin 6
  • Filter board (A8, J3-P3) pin 10
  • Filter board (A8, J10-P10) pin 10
  • Knocker connector (A7, J1-P1) pin 2
  • Knocker connector (A7, J1-P1) pin 1
  • Filter board (A8, J10-P10) pin 9
  • Filter board (A8, J3-P13) pin 9
  • Main PCB (A1, J1-P1) pin 6

At every location I could see +30v DC so was a bit stumped at this point as to why things weren’t working. What I decided to do next was to verify that everything worked beyond the PCB, and thus proving that the PCB had a fault, so what I did was ground the tab of the 2N6044 transistor @ Q2 on the main PCB, and nearly got the shock of my life when then coil activated and the knocker fired. This confirmed that the fault was on my main PCB.

I didn’t know if the fault was now a logic one, or an issue with the transistor at Q2. So I hooked up my logic probe to the main PCB and attached it to the 74LS377 @ A8, pin 15. This is the chip which sends a signal to the transistor in order to activate the knocker. I then played the game and watched my logic probe as Q*bert fell of the pyramid. I could see that the logic changed from a low state, to a high state confirming that it was trying to activate the coil and knocker.

This then led me to believe that the transistor @ Q2 had failed and this was where the problem was. Unfortunately I did not have any spare 2N6044 transistors but noticed that the same transistor existed on the PCB @ Q1 and Q4 and these were optional transistors for coin 1 and coin lockout, so I used one of these to replace Q2. I also confirmed with a tester that Q2 had failed, and it had:

Once the new transistor was in place, I put the main PCB in the cabinet, powered up and the game, and was immediately greeted with the sound of the knocker. I then played a game, and when Q*bert fell off the pyramid was delighted to hear the sound of him hitting the bottom of the cabinet. Knocker fixed!