Announcement

Collapse
No announcement yet.

Donkey Kong TKG4 Repair Logs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Donkey Kong TKG4 Repair Logs

    I've got a stack of DK TKG4 boards to go through and fix so I'll post the repair logs here as I go.
    All boards are tested with a known good other layer, I'll start with the CPU PCBs first then move onto the Video PCBs.

    CPU Board
    Fault: No Digital Sounds

    The board was working fine except for no digital sounds.
    First thing to do was dump and check the ROMs which were fine and I swapped out the 8035 sound chip which was also working fine.
    In the video below you can hear the analogue sounds working, near the end of the video you can hear a static noise which was caused by touching the AD4 address line on the ROMs at 3F & 3H (pin4) with the logic probe.

    https://youtu.be/xC7-2X4P_Os

    The address line lead back to a 74LS75 at 4F, it was removed and tested faulty.



    I put a new one in and all digital sounds were restored.
    Last edited by Kaizen; 29th May 2019, 10:41 PM.
    Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

  • #2
    CPU Board
    Fault: Garbage on screen/resetting - No jump sound - Graphics glitch on right side of screen.

    This board would just show garbage on screen, when I tested it 9 months ago it would show garbage on screen and constantly reset, now it's just static garbage.



    After going over the board I couldn't find any issues with the TTL chips that I looked at, the Z80 was tested and working fine but there was no activity from the CPU, the 8257 DMA controller wasn't socketed so it was removed, socketed and a good one fitted back in.
    The board now boots to a working game but there was a graphics glitch on the right side of the screen and the jump sound was missing.

    https://youtu.be/eSvx97TcKxU

    On closer inspection of the jump sound circuit I found one of the transistors had a broken leg.



    It was replaced and the jump sound was restored.

    When pressing down on the P2 ribbon cable connector that goes between the two layers, the graphics glitch would disappear. Reflowing the connector fixed that issue.



    The board is now fully working.
    Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

    Comment


    • #3
      CPU Board
      Fault: Garbled graphics

      This was a relatively easy one, the game would boot to garbled graphics and run for a few seconds then freeze.

      https://youtu.be/hGBATuKRh7Q

      I started looking around the CPU area
      and found the outputs coming from the 74LS74 at 7F were all dead.



      The chip was removed and replaced and the board now boots up and plays normally.



      CPU Board
      Faults: Wouldn't sync - Only green colour - Missing Kong thump sound - no 'Up' input from the joystick

      The board had copped a decent hit at some stage and there were two transistors, a number of resistors and a cap missing/damaged on the final video stage.



      These were all replaced and the colours and sync were restored.

      Right near the edge connector, the board had had an impact and the 'Up" input signal trace had been cut. A resistor leg was inserted into the via, laid over the broken trace and soldered in place.



      To check the pounding noise that Kong makes at the start of the game when knocking down the girders, you can short the Collector to the Base of Q2 (or touch the base with a logic probe) and you should hear the noise, when I did this there was nothing. I checked the transistors and caps which were all good. I followed the base of Q2 back to the 74LS161 at 3J and found pin 12 was stuck low.



      After replacing it the Thump sound was restored.

      The board is now fully working.


      CPU Board
      Fault: Boots then screen goes blank


      https://youtu.be/FY2oSl7C_M0

      This board would boot up then withing a few seconds the screen would go blank.
      I checked the reset line which was good and worked my way down the left side of the board and when checking the 74LS139 at 2A, the outputs were toggling.



      Replacing it fixed the issue and the board is now fully working.
      Last edited by Kaizen; 7th June 2019, 07:43 PM.
      Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

      Comment


      • #4
        After getting six CPU boards working, I wanted to make them reliable as possible.
        These boards had a lot of chips with corroded pins and most of them were Fujitsu's which I planned on replacing anyway.



        Every Fujitsu that I pulled had one or more legs hanging on by a thread. The bent pins in the image below are the ones that had no structural integrity and bent or broke off when lightly touched.
        15 Fujitsu's with almost 60 shagged pins. Surprisingly they were still working.



        All the original single wipe sockets will be removed and replaced on all boards.
        Here's the results of the last couple of days' efforts, a total of approximately 1300 pins that had to be cleared.

        Last edited by Kaizen; 1st June 2019, 09:28 PM.
        Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

        Comment


        • #5
          CPU Board
          Fault - Garbage on Screen/No Activity



          This board was showing almost exactly the same screen as in post #2 .

          The 8257 DMA Controller was soldered in so it was removed, socketed and a known good one fitted.
          The board would now boot up but was constantly resetting.

          I checked over most of the board and all the TTL's looked to be behaving, I had a feeling this was RAM related and one of the other boards had the RAM socketed so I pulled all six of them out and tested them one by one and found two of them to be faulty.
          I soldered the good ones back in and replaced the two faulty ones with known good ones.
          The board is now fully working.

          CPU Board
          Fault - Game plays with corrupted graphics


          The game would boot up and play but the graphics were corrupted.

          https://youtu.be/B_ytq0Z4N_E


          On closer inspection of the board I could see it had taken a hit near the interconnect.



          The trace in the middle was completely cut and the other two were damaged but still had continuity.



          Some Kynar wire was laid over the cuts and soldered in place.



          The board would now boot up and play but the walk sound was wrong.

          https://youtu.be/J0CmnvnFH8I

          Starting from the 74LS05 inverter at 6L I followed the signal to the first cap (C25 - 3.3uF), it was pulled and tested bad, almost 6uF.



          After replacing it the walk sound was restored to normal and the board is now fully working.

          That's the last of the CPU boards, I'll start on the Video PCBs now, there's eight of them with various issues from no image on screen to corrupted graphics.
          Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

          Comment


          • #6
            Video Board
            Fault - Corrupted Graphics


            The board would boot up and play but the graphics were corrupted.

            https://youtu.be/A7kE_3_AoPo

            I replaced all the factory fitted single wipe sockets, someone had previously removed the two 2114 RAM at 2P and 2R and fitted sockets.



            On closer inspection I could see that one of the traces had been lifted and there was no continuity between the RAM and the PROM that the trace leads to.



            The PROM was checked and it was found to be faulty, the RAM was also checked and they were good.

            I ran some Kynar wire between the pins and powered the board up again and the game now runs correctly.



            I'm guessing that the PROM was faulty and the previous 'tech'had thought it was the RAM, socketed and replaced it but caused another fault in the process and gave up on the the repair.
            The RAM sockets will also get replaced and I'll try to repair the trace at the break and remove the Kynar wire on the bottom of the board to make the repair a little cleaner.


            CPU Board
            Fault - Corrupted Graphics - Wouldn't Sync


            This board had been bashed around and had broken sockets, broken H & V position pots and a heap of bent pins on the under side of the board.







            After resocketing the board, straightening and reflowing the bent pins and fitting some 15k resistors in place of the pots (temporary solution) the board powered up and ran fine.
            That was an easy one...

            There's a heap of Fujitsu chips on this board and some of the other boards as well, these will all get replaced.
            Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

            Comment


            • #7
              Video Board
              Fault - No image on screen


              The H and V position pots were missing and the trace leading to the V pot had been lifted off the board.
              15k resistors were soldered in place of the pots, one where the Horizontal pot goes and the other to where the Vertical trace leads to, the other end of the resistor was tied to the +5v on the chip next to it.



              After power up I did a quick check of the ROMs which showed the board was dead in the water with no activity whatsoever.
              I went straight to the crystal and checked it which also had no activity.
              You can see in the image above that the copper shield had taken a hit and is squashed down.
              When the shield was removed I could see both transistors had been squashed and a continuity check showed that both had shorted pins.





              The legs were straightened and both transistors tested okay so they were put back in, the shield was also straightened and refitted.



              After powering the board back up it was fully working.
              The lifted trace was still intact, I've straightened it and it will get glued onto the board.
              Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

              Comment


              • #8
                Video Board
                Fault - No image on screen


                The board had broken V and H position pots so they were replaced with 15k resistors for testing.
                Same as before, a quick look at the ROMs told me there was no activity and the board had no clock signal.
                I threw the logic probe on the crystal and it was humming away nicely.
                The crystal frequency is 61.440MHz, this feed into a 10116 ECL chip (different to TTL) which can't be checked with a logic probe so I fired up the Oscilloscope to continue with the troubleshooting.



                The input to pins 12 and 13 at 1D (10116) were sitting at 61.4437MHz so all looked good there.



                The output at pin 14 however wasn't doing much at all.



                1D was replaced and I had a good signal on the outputs.






                The output from pin 6 at 1D goes to pin 13 on the 10136 at 1E, checking the outputs of 1E at pins 2 and 15 showed they were lifeless.



                Replacing it gave me 7.6MHz at pin2 and 15.3MHz at pin15.




                When checking a working board I could see that both pins should have around 12Mhz (somewhere in the middle of these values).
                The output from pin2 feeds back into pin7 via the 10105 at 1F.
                With this being the only other part of the circuit I deduced that it was faulty and swapped it out, I now have 12.3MHz on both pins 2 and 15 at 1E.



                With 1E and 1F now working correctly I have good inputs of 12Mhz going to 1H.

                The outputs at 1H however were all dead (I see a pattern forming here).



                1H was replaced and I now had 12MHz coming out the other side and there was now a corrupted image on screen.



                https://youtu.be/8IuwWRoeYVw

                A quick look at the other ECL chips on the board showed that they had good inputs and bad outputs so I'm guessing they're also shagged.

                To be continued...
                Last edited by Kaizen; 6th June 2019, 01:51 PM.
                Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

                Comment


                • #9
                  ECL runs from -5v and if they are all dead then there was either a catastrophic failure in the power supply or the wrong voltage was connected to the -5v supply to the board.

                  Comment


                  • #10
                    Yeah I was thinking the latter...

                    Video Board
                    Fault - No image on screen - Corrupted graphics

                    The initial fault with this board was that it was not displaying any video.

                    When checking the voltages at the chips I found the -5v was sitting at -4.2v.
                    Cleaning and reflowing the 10 pin rainbow connector got the board displaying an image and starting a game but the video was corrupted.



                    When looking at the data lines on the 74LS245 at 1S, I found pin5 was stuck low.



                    Replacing it fixed the issue and the video now displays correctly.
                    Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

                    Comment


                    • #11
                      Bloody awesome effort. Well done.
                      This is where it ends!

                      Comment


                      • #12
                        Originally posted by Kaizen View Post
                        CPU Board
                        Fault: Garbled graphics

                        This was a relatively easy one, the game would boot to garbled graphics and run for a few seconds then freeze.

                        https://youtu.be/hGBATuKRh7Q

                        I started looking around the CPU area
                        and found the outputs coming from the 74LS74 at 7F were all dead.

                        https://i.imgur.com/Qoo7Wri.jpg

                        The chip was removed and replaced and the board now boots up and plays normally.



                        CPU Board
                        Faults: Wouldn't sync - Only green colour - Missing Kong thump sound - no 'Up' input from the joystick

                        The board had copped a decent hit at some stage and there were two transistors, a number of resistors and a cap missing/damaged on the final video stage.

                        https://i.imgur.com/N9Frnig.jpg

                        These were all replaced and the colours and sync were restored.

                        Right near the edge connector, the board had had an impact and the 'Up" input signal trace had been cut. A resistor leg was inserted into the via, laid over the broken trace and soldered in place.

                        https://i.imgur.com/cjGcNE5.jpg

                        To check the pounding noise that Kong makes at the start of the game when knocking down the girders, you can short the Collector to the Base of Q2 (or touch the base with a logic probe) and you should hear the noise, when I did this there was nothing. I checked the transistors and caps which were all good. I followed the base of Q2 back to the 74LS161 at 3J and found pin 12 was stuck low.

                        https://i.imgur.com/IAk7yOY.jpg

                        After replacing it the Thump sound was restored.

                        The board is now fully working.


                        CPU Board
                        Fault: Boots then screen goes blank


                        https://youtu.be/FY2oSl7C_M0

                        This board would boot up then withing a few seconds the screen would go blank.
                        I checked the reset line which was good and worked my way down the left side of the board and when checking the 74LS139 at 2A, the outputs were toggling.

                        https://i.imgur.com/Luio3ev.jpg

                        Replacing it fixed the issue and the board is now fully working.
                        Hey. I think I have chased down the same problem on my board so I will be looking at the equivalent LS139 chip on my board tomorrow (2C instead of 2A). The outputs should be pulsing right??

                        Comment


                        • #13
                          Originally posted by hvc01 View Post
                          Hey. I think I have chased down the same problem on my board so I will be looking at the equivalent LS139 chip on my board tomorrow (2C instead of 2A). The outputs should be pulsing right??
                          What's pin 1 doing?
                          Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

                          Comment


                          • #14
                            Originally posted by Kaizen View Post
                            What's pin 1 doing?
                            all pulsing apart from grounded pin. i then checked 4H (TKG4 4D) and inputs pulsing but output pins 12,13,14 stuck high. replaced. no change. time for bed now.

                            i would next be going to 8F and 7F to check here.

                            Comment


                            • #15
                              Video Board
                              Fault - Graphics Glitch Above The Top Ladder




                              The glitch only appears on the barrel screen when the sprites are moving.

                              I've had this issue on a couple of boards and turning the voltage up reduces it but it won't go away until the +5v is at 5.25 volts (not good for TTL chips) and -5v is around -5.00 volts (usually sits around -4.74 when +5v is at 5.00volts).
                              Initially I though it was related to the voltage drop between the two layers and later thought it might be related to the ECL chips but swapping them out on one of the boards made no difference.
                              I'd spent a few hours on this issue and it was doing my head in.
                              I put it in the too hard bin for a few months and decided to take another look at it tonight.

                              After going over the ECL chips again with no luck I decided to go for a random walk around the board with the logic probe and see if I could spot anything that didn't look right.
                              I checked all the lower order TTL chips and all looked good then I came to the 74LS373 Transparent Latch at 3K and started to look at the inputs/outputs and when I touched pin 18 with the logic probe I found the glitch would disappear.
                              I pulled the chip and it tested okay, 3K was then socketed and and I put a new chip in with no change.
                              I had a similar thing happen on a Pong board where the capacitance of the probe (or my finger) was enough to fix the issue.

                              Putting the original chip back in and adding a 22pF ceramic cap across pin 18 and ground fixed the issue (piggybacking it with another 74LS373 also made it disappear).
                              Namco Starblade - Taito Cocktail 14" x 3, 20" x 2 - NZ SI PtII Upright - Chase HQ DX, Taito Top Speed - Taito Grand Champion - Taito Safari - LAI SI Head to Head Cocktail - SI Cocktail Clone x 2, Pong Clone x 3

                              Comment

                              Users Viewing Topic: 0 members and 1 (guests)
                              Working...
                              X