1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

#15 Twilight Zone w/ Lots of Issues

Discussion in 'The Menace Files' started by Menace, Jun 13, 2016.

  1. Menace

    Menace Well-Known Member
    Staff Member

    Joined:
    Nov 14, 2012
    Messages:
    2,424
    Likes Received:
    221
    Location:
    Santiago de Aurora
    The latest installment to my repair journey had me flexing my troubleshooting skills to the max.

    One of my clients contacted me months ago about issues with his TZ. Turns out his display was showing nothing but garbage after swapping in a ColorDMD. On top of that, the game was doing a bunch of other weird things like firing all the coils momentarily when you powered the game up, the clock was throwing "Clock Broken" errors and the flippers were so weak that when you pressed the flipper buttons they would barely make a full flip! I told him that I unfortunately didn't have time in my schedule to get up to his place, plus with the amount of issues the game was having I felt it would be best to have it in my shop to troubleshoot. Fast forward a few months and I finally had enough room in my shop to make room for delivery, so the owner brought it down.

    Now, between the time he initially called me for help and the time I got my hands on the game another tech had gone through the game at his home, and unfortunately was not able to determine the cause of all the issues. This is what I was dealing with on power-up



    As you can see, that just isn't right. On top of the garbled display, I had the "Clock Broken" error, I had GI in the backbox and sections of the PF slowly strobing, and in test when the flipper buttons were pressed the flippers were SUPER weak and would barely make a full flip. I decided to start with the display, as that is crutial to testing everyting else in the game.

    Initially it would appear that the data from the game being sent to the display was corrupt, and the first thing you check for that is ribbon cables as it's the easiest thing to do. (all of the other issues the game had wouldn't normally be assosiated with ribbon cables) At some point all of the ribbon cables in this game had already been replaced with repro units, so the chances of it being a bad cable was remote but I did find this...

    tz1.jpg

    As you can see this connector was not installed correctly! This is the daisy chained cable from the MPU to the Fliptronic's PCB over to the sound PCB and then finally the DMD controller PCB. Unfortunately for me this didn't fix ANY of the issues with the display or anything else happening in the game. The next thing I tried, again just because it was easy, was swap ALL of the cables from the game out with known working ones from another game. Once again, no change. Where the game wasn't using a stock plasma DMD I wasn't sure if the display issue was with the ColorDMD assembly or not, so I swapped the ColorDMD with a stock plasma DMD.

    At this point I made a little progress, as the stock DMD was finally showing correct display information but there was still an issue with it. If anyone has seen an outgassing DMD, that is the best way to describe it but I know the DMD is perfect as it came from one of my own games. On power up the display was only showing informatoin sporradically, mostly on the left side of the DMD but the longer I had the game powered up the better the image would get to the point where in test I could finally see all of the menus. Where I knew the DMD was good I suspected a failing High Voltage power supply section on the DMD controller PCB but probing it with my DMM all of the voltages were well within spec! WTF?

    What we know is the DMD is good, the voltages are good and all of the ribbon cables are good. I have a spare working DMD controller PCB so I swap that into the game to eliminate that from the equation, and you guessed it... no change! So at this point I have pretty much eliminated all of the easy stuff and I need to turn my attention further up-stream.

    With the DMD working enough I figured it was time to put the game through some of it's self tests. First I ran the DMD tests, which passed as I suspected it would at this point. I ran the coil tests and noticed that a bunch of the coils were not working even though all the fuses are tested good. I ran the clock test, which presented a little more information. The clock was actually working, but the clock hands were running REALLY REALLY slow which is why the game kept throwing the "Clock Broken" error. It was as if something in the game was tanking all of the 12VDC power driving the clock motor along with all of the opto's in the game. I had a bit of an A-HA moment at this point, because remember the weak flippers? TZ uses flipper opto PCB's so I realized that the flippers and the clock issue are now related, but I still didn't know how. Where all of these issues seemed to stem from the DMD, for giggles I ended up disconnecting the ribbon cable to the DMD during the clock test and wouldn't you know it the clock hands began to move at their correct speed?!? I then hit the flipper buttons and those too were now working perfect! Reconnect the DMD ribbon cable and the clock and the flippers both sh*t the bed again. :banghead:
     
    #1
    Last edited: Jun 13, 2016
    WARLOCK likes this.
  2. Menace

    Menace Well-Known Member
    Staff Member

    Joined:
    Nov 14, 2012
    Messages:
    2,424
    Likes Received:
    221
    Location:
    Santiago de Aurora
    So I pulled the driver PCB just to give it a good visual front and back, and came across this....

    tz2.jpg

    This is a blown out trace on the A/C input connector for the 12V power supply section. It would appear at some point in this game's life there was enough of a current draw to destroy this trace. (from what is anyone's guess but I have my suspicions which I will get to later) The reason the game was actually still working is this input connector is doubled up and you can see the other trace is still intact. I repaired that trace with a jumper wire and toss it back into the game.... and still the same issues. W-T-F?! It's at this point I'm still trying to determine where exactly the issue is. Before I jump to my logic probe I figured I would pull the driver AND the MPU, and I would test both of these in my own TZ to try and eliminate as many things as I can from the equation. I'm still not sure if the issue I'm dealing with is actually on any of the PCB's in the head or if it's something else in the game causing the problem.

    Now one thing I would like to stress to those that swap PCB's between games for testing purposes... if you are going to attempt something like this you understand there is a certain amount of risk involved. ALWAYS ALWAYS ALWAYS move the suspect PCB into a known working game and not a known working PCB into a suspect game! Reason being, if you move a known working PCB into a suspect game (specifically a driver or MPU) you run the risk of an issue in the game destroying the good PCB! If you have a suspect PCB and you install it into a known working game the chances of ruining something in the game is much much less, and if something locks on you should notice right away and turn the power off before anything bad happens.

    I knew well enough at this point nothing on the driver or mpu would ruin anything in my game, so I started with the MPU. Swapped them and flicked the power switch the DMD fired right up, as did the clock and my flippers and all the coils etc. Perfect, one more PCB eliminated from the chain. Swap in the driver and fire up the game, and my TZ shows a messed up DMD, slow clock and flippers and a bunch of the coils aren't working. I now know that there is something on the driver PCB causing all of the issues. I disconnect just the ribbon cable between the MPU and the driver and power the game back on and the display comes up perfect! Where the ribbon cable to the driver is disconnected (the data bus) I can't test any coils or lamps but this tells me it is likely something to do with the driver data bus that is dragging everything down.

    So I grab the schematics and study the data bus section of the driver PCB. There are a chain of IC's on the data bus that process info coming in from the MPU (74LS374N), each one is tied to a specific section of the driver (GI, inserts, coils etc) Using my DMM in diode test you are able to test these IC's pretty easily by putting the red probe on ground and the black probe on each leg of these IC's, and you should get 0.4 to 0.6 for all but two legs. Sure enough the very first IC in the chain that controlls the GI was showing bad readings, but the rest were ok. So I desoldered the U1 chip, installed a socket and popped in a new IC.

    tz3.jpg

    Toss the driver back into the game, and the DMD is now working 100% as are the clock, the flippers AND all of the coils! :woot: I was amazed at how one IC on the databus that is responsible for controlling the GI could fail like this and tank the entire data-bus in the game. I proceed to play a couple games just to make sure everything was working and it played like a champ, so I tossed the ColorDMD back in and it too is now working as it should.

    But before I was about to call it a day I noticed that NONE of the GI was working, completely dead! Check fuses, all OK. How could a brand new chip that is responsible for the GI not be working right? Toss the game into GI test and this is where it got weird. If I ran the All GI test, nothing would light up. If I ran the separate GI section tests, each section would work no problem (upper right, upper left, lower right etc etc) But if the game tried to light ALL of the GI sectoins, nothing. This was going to call for a logic probe.
     
    #2
    Last edited: Jun 13, 2016
    WARLOCK likes this.
  3. Menace

    Menace Well-Known Member
    Staff Member

    Joined:
    Nov 14, 2012
    Messages:
    2,424
    Likes Received:
    221
    Location:
    Santiago de Aurora
    Now I'm not going to get into how to use a logic probe because it's WAY too involved, and the information you gather using the probe will depend on what IC's you're probing etc. The really short version of this story is I managed to figure out that some of the data being sent to the new IC I had just installed was inverted, which means the data being sent was wrong. Another quick look at the schematics showed me that the GI data coming out of the ASIC on the MPU travelled through one IC (74LS240N @ U12) before coming over to the driver PCB, and I confirmed the data coming out of the ASIC was correct. This led me to believe that U12 IC was bad so I pulled that chip on the MPU, installed a socket along with a new IC.

    tz4.jpg

    My troubleshooting was correct, and this solved the GI issue I was having. With the game now 100% working I logged a number of test games to ensure no other issues were present.

    When the owner of the game brought it down to me, he also asked me to install a couple more mod's (to an already seriously modded TZ). Here are a before and after pic of the GLM mini-PF switch kit and a lazer cut lamp shade.

    Before

    tz5.jpg


    After

    tz6.jpg

    The GLM kit is probably one of my favorite as it cleans up the look of the mini-PF.

    Now earlier in this repair I touched on the blown out trace from the 12VDC input power connector. Given the number of mods in this game and the fact most of them rely on the 12VDC as a power source, I have long suspected that the stock driver 12V section cannot handle the amount of power required for the game AND the mods people are installing. I say this because it's a known fact that people with STTNG who install ColorDMD's (which rely solely on 12V to power it) have nothing but issues, which is not surprising given the number of opto's used in that game... they are all powered from the 12V supply. My guess is this TZ had one too many mod's installed and the traces on the PCB were not designed to handle the amount of current everything required to run and burned the trace up. (either that or there was a short circuit condition at one time with the wrong fuse installed) Here's a quick shot of just SOME of the mod's connected to the driver in this game, keeping in mind there are at least 2 or 3 that I didnt' connect! I asked the owner if he was related to Clarke Griswold?

    tz7.jpg

    Anyway, this was a long one and it took me a while to track down the actual cause because the issue was masked by so many other potential problems. Certainly not an easy one to diagnose, but I learned a few things along the way and the owner is just happy to be able to play his TZ again after it being off-line for so long.

    D

    PS. The owner of this game is a good friend, and we like to joke around with him A LOT. Before I folded and wrapped the game for pick-up I decided to mess with him just a little bit. :woot:

    tz8.jpg
     
    #3
    Last edited: Jun 13, 2016
    WARLOCK, dnewman, tkaye and 2 others like this.
  4. tkaye

    tkaye Active Member

    Joined:
    Mar 17, 2013
    Messages:
    510
    Likes Received:
    27
    Location:
    Ancaster, On
    Great post Doug! Always amazed at your never ending diagnosis talents! Another mystery solved and job well done!
     
    #4
    Menace likes this.
  5. Chris Bardon

    Chris Bardon Well-Known Member

    Joined:
    Nov 15, 2012
    Messages:
    1,060
    Likes Received:
    62
    Location:
    Mississauga, ON
    Thanks for the write up-interesting to read the process that goes into diagnosing something like this. Way beyond where I'm at with this stuff.
     
    #5
  6. Bigbossfan

    Bigbossfan Active Member

    Joined:
    Aug 22, 2013
    Messages:
    443
    Likes Received:
    81
    Location:
    Wasaga Beach, ON
    Hahaha...........yes I am related to Clarke Griswold. A few notes. In regards to that misaligned ribbon connector. At one point, that was the only way that Astill and I could actually get the game to play........f'n Gremlins. Then of course it stopped working all together.

    I am amazed though after reading this all that you had to go through to figure this out. As I've told you before bud, I'm very impressed with your skills, you're a good man and thanks for getting my TZ up and running after being down for about 3 months. Now that I've been nice..........how about you eat some of these??

    bag.png
     
    #6
    Menace likes this.

Share This Page