Not a real repair but just a quick “dirty” fix for this Konami Teenage Mutant Hero Turtles PCB :
All was fine but the music played at the start of a game (when turtles jump from palace) was missing :
A usual I started my troubleshooting with a visual inspection and I found soon the culprit.One of the two 640KHz ceramic resonator was missing:
I took it from a dead donor board :
and once fitted the missing music was restored.End of (small…) job.
I had a spare Sega System16B ROM board type ‘171-5797’.Originally it mounted the E-Swat ROM set but I decided to convert to Golden Axe since I missed this game in my collection :
Atter burning the Golden Axe MAME rom set, I powered the board up and game was fully playable but with a noticeable sprites issue:
Judging from fault, I immediately thought about an addressing trouble of the sprite EPROMs (27C020 devices) so I went to check for continuity of their address lines.They were properly daisy-chained each other but the pin 30 (A17) of all of them was not connected to any pin of the ‘CN2’ interconnect (signal to address lines come from motherboard) :
Lack of signal on this A17 address line was confirmed also by my analog scope (good signal on left, bad on right of the picture below):
So, the trace had to be broken under the ‘CN2’ interconnect.Being not available schematics and wanting to avoid to desolder the whole connector, I soldered the end of a wire to PIN 30 of a sprite EPROM and with the other end I touched each pin of the ‘CN2’ interconnect that gave me no continuity until I hit the right one restoring correct sprites:
Lastly, I made the fix permament holding the wire with some hot glue:
That’s all for today.
Some days ago I bought on Ebay for cheap an untested original Sega Pengo PCB, today it arrived:
Buying a board as untested is always risky, you could ended up to have a not working one on hand and indeed mine was.I was greeted by a solid black screen when I powered it up, the main Z80 CPU didn’t reset properly.As schematics suggested the /RESET is generated by a counter 74LS161 @IC23 so I went to probe it and found it was missing clock signal.At this point I started to suspect some trouble in the timing signals generation.The Pengo manual has a troubleshooting section which suggests to check some iC in case the picture will not appear:
When I went to probe a 74LS74 @IC66 :
I found its outputs pin 5 and 6 were stuck high.I put a my fingers on it and it was really burning hot.Also comparing it against a good one with my HP10529A logic comparator confirmed my suspicions:
Once removed, the IC failed the out-of-circuit test:
Fitted a good IC fixed the board completely, no further issues were found.
I had this rare Irem Ken-Go PCB (an undumped version included in latest MAME release) from my friend Joachim for a repair:
The board played fine except for an issue where all sprites were formed but missing all colors appearing like black shape:
Sprites generation circuit is located on top board where some customs lie:
Documentation for this specific game was not available but luckily I could know function of the various customs by looking at the R-Type schematics and so identity the one involved in the sprites palette generation:
The custom itself was marked ‘KNA91H014’ in 60 pins QFP package:
Probing it revelead that inputs were fine but outputs had irregular signals:
So I opted for replacing this custom (taking a good one from a dead Vigilante board) :
This was the right move since sprites were correcly restored:
End of job.
Got this Konami Haunted Castle PCB for a repair:
When I powered it up for first time, two were the faults I noticed : sound and sprites were missing.So I started to troubleshoot first issue.The sound circuitry of this hardware is made of an Z80 CPU,a 6116 SRAM, a 27256 ROM and YM3812 DAC plus a couple of custom for PCM samples and music generation (K007232 and ‘K051649’).When I went to probe the Z80, I found that most of data lines were stuck LOW or HIGH.The ROM containing sound code was good as well as the relative RAM.With the help of schematics I figured out that sound code data were latched by a 74LS374 @G11 :
Probing its outpus revealed they were stuck and piggybacking it caused playing some random samples.So I decided to remove the IC :
It failed when tested in my programmer:
Sound was fine but, as I said, sprites were missing (except for some):
Since I already repaired in the past a couple of Haunted Castle boards with identical issue, I knew where to look at.Hardware uses for sprites and tiles some 4464 DRAMs in ZIP package.In particular the two related to sprites are the ones @H3 and @H4:
So I ordered some spare parts from China.When they finally arrived , I could successfully finish my job by replacing them:
Another great game preserved!