Tuesday, 21 August 2018

Bagman pcb repair notes.


I've labeled this one as notes rather than a repair log as it was three different boards I repaired at once and not everything is covered. A lot of TTL needed to be replaced to get these up and running.

Some issues I noticed across all three boards were that the Z80 cpus were bad and when removing them from their socket some of the pins broke off and stayed in the socket.

Missing player / enemy sprites on two of the boards was down to a bad 74LS163 @ M2
Can also be caused by a bad 74LS32 @ 2P

No speech (TMS5110A was ok) on one board it was a bad PROM @ R6 and on the other a bad 74LS151 @ S6

Bad PROMS in general, on one of the boards all three were bad. This causes colour issues and no speech.

An issue across all three boards was bad / missing video sync. This was a combination of bad 74LS161s @ A4, A5, B4 and B5.

Bad 74LS283s @D2 and D3 again on all three boards.

Certain controls not working 74LS245 @ 9A or 2B

The four sprite rams (27LS00) were fine on all three boards.

One final note, the PAL @ P6 is used for 'guards controlling' according to mame, which I guess is a cryptic way of saying controls the enemies. With the PAL removed the enemies seem to behave the same but I am guessing later in the game they behave differently and that it might make the game harder with it removed.

A similar security measure to what Bubble Bobble used maybe?

Download the full size image here

To help with my repairs I took the existing pdf scans of the schematics which are split up over several pages and combined them into a single page using photoshop. There's still quite a few areas that aren't legible unfortunately but I don't think the originals were that good to begin with.

Monday, 6 August 2018

Final Fight pcb repair

This original CPS1 Final Fight pcb was working fine apart from no sound.

The most likely causes were the Z80 cpu (capcom fitted Z80 instead of Z80A), 6116 work ram for the Z80, sound eprom on the B board and less likely but also possible the amplifier or oscillator.

I started with the sound eprom for some reason this one always seems to be missing a label on many CPS1 games which lets the bad UV light in. This was fine though, next I had a close look at the board it was very clean and looked like it had been well looked after.

The 6116 ram on the other hand had corrosion on several of legs, I have read that some early 2K X 8 rams contain electrolyte which eventually leaks. I am not sure if this is true but it would explain why these Sony rams often fail and the corrosion on the legs which was no where else on the board.

With a socket and new old stock compatible ram fitted the sound was fully restored.

Saturday, 2 June 2018

Sunset Riders pcb repair

This was just sold as faulty so I wasn't sure what to expect with this one. First of all a visual inspection, I noticed right away a deep scratch on the bottom of the board that had taken out several traces. Someone had taken off some soldermask and figured out which were the bad traces but not made any attempt to repair them.

I was going to use kynar wire but this was too big, the traces are extremely small and close together. So I ended up taking some multi core wire and using a single strand about the size of a hair.

With the micro surgery complete and no other visible issues I powered the board up. At first all I got was a black screen so didn't think it was running but then the music kicked in and part of the title screen appeared.

I fixed another Sunset Riders pcb recently that didn't end up getting logged but it gave me a good idea where the issue was. The other board wasn't as bad but had colour and background issues caused by a bad 053251 konami custom.

By touching certain pins on the custom chip I was able to get it to display slightly more of the background.

I took a donor 053251 from a Golfing Greats pcb.

Not quite there yet but better.

Time to check the maskroms as these are another common failure on this era konami board. I went into test mode but wasn't able to read which was the maskrom test, I used mame to see what I should be seeing. Then I could see the bottom two turned red which again in mame I could see was 16K and 1D, 16K maskrom contains graphics data and D1 contains the audio. This explained the lack of shooting and certain other sounds.

I then replaced the two maskroms with eproms (27C400 and 27C800). Luckily they were all in sockets from the factory, usually they're soldered directly into the board.

A blob of liquid soldermask to protect the repaired traces on the bottom and the repair is complete.

Tuesday, 29 May 2018

Popeye pcb (bootleg) repair

I bought this Popeye bootleg pcb in a lot of faulty boards, it included the worse jamma adapter I've ever seen. It's made from another cut up pcb, anyone know what game it was?

The board was working apart from the character layer of the graphics, making the text scrambled and blue blocks appear over the background and other sprites.

I didn't notice at first but blue and green were also switched, this was an issue with the adapter though and not the board.

The first thing I did was dump all the eproms and make sure they matched the ones in mame. Eprom 5 failed but this was a red herring as it was fine even though it's not in mame. :)

This did lead me in the right direction though as the data causing the issue is stored on this rom, it looked much better with it removed from the board. So I started looking around this area, I piggybacked a 2114 ram @ F8 and things looked much better, leaving the ram in place and power cycling the board completely fixed the issue.

I replaced the bad ram and made a new jamma adapter.

The colours are now correct, repair complete.

Saturday, 26 May 2018

Aliens pcb repair

I bought this Aliens pcb as faulty with the error bad i3.

Someone had already replaced and socketed i3 (a 6116 type ram) and this made no difference to the error.

Checking the very nice schematics there was pretty much only one thing that could be the issue.

The 051960 konami custom chip. I had one left in my little box of customs, in unknown working condition. It came from a Main Event board that I'll be harvesting for more parts later in the repair.

Anyway back to the 051960

Replacement custom in place.

Yay the ASIC is good.

This just left the sound section, four capacitors were broken off the board.

The electrolytics I have plenty of but I didn't have the correct value for the mylars so I dug out the scrap Main Events board and pulled two from that.

Sound fully restored, repair complete.

Saturday, 21 April 2018

Thunder Force AC repair

After a long break from board repair I finally got the ball rolling again with this Thunder Force AC conversion pcb I just picked up. It was sold as having issues with the background graphics and that was exactly how it behaved for me too. The background would constantly flicker, everything else seemed fine.

On close inspection of the pcb I found a few issues, there were a couple of small patches of corrosion. Most of it just needed cleaning but one spot it had gotten as far as eating through one of the traces.

I tinned all the traces in the area where there was corrosion and added a tiny bit of wire to the broken trace.

Then finished it off with a small amount of liquid solder mask. This fixed the background issue, all that was left to do was replace a broken resistor.

Repair complete.

Sunday, 8 October 2017

Spy Hunter pcb repair

I wasn't planning to log this one but as the fault turned out to be something I haven't come across before I thought it was worth mentioning. That's also why the screen shots are from mame.

This playtronic Spy Hunter pcb would either start with no background graphics or run fine for an hour or two and then the background would vanish.

Intermittent faults are always time consuming and a lot of guess work, it looked like a failing chip that was failing as it heated up but often from cold there was no background and spraying chips in the background circuit with freeze spray made no difference.

The midway schematics came in very handy, the playtronic board doesn't have screened chip locations but the schematics include a picture of the pcb with locations. After ruling out all the single wipe sockets for the ram (2b) and eproms (3a - 6a) used in the background I was left scratching my head.

I had previously replaced all the ribbon cables and also reflowed the connectors with no change to the fault.

I took a short break and when I came back to the board the background had returned, this makes finding the fault impossible so I had to leave it on for around an hour for it to vanish again.

Once the background vanished I checked all the outputs of the TTL connected to eproms (3a - 6a).

The outputs on the 74LS174 @ 13e were all low, the truth table for the 174 and schematics confirmed that pin 1 should be high as it's connected to +5v through a 1K resistor but it was reading low.

Checking the resistor and 5v was going in one side but nothing coming out the other, I changed the resistor and sure enough 5v back at pin 1. The resistor was obviously going open but not just from heat as from cold it was often open too. The resistor has no signs of damage or corrosion.

Best of all out of circuit the multi meter says that the resistor is fine and exactly 1K.