Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by KAISERSIGMAX3

  1. Wife and kids were out, so I went to play some 1944 - The Loop Master (personally I like 19XX a lot more!). Near the end of the game the system crashed and displayed that "Error Exception Handling" black screen. So I turned off the system and on again waiting for the worst.... Humph, for the next 5 tries I was given a blue screen.. In other words, the system was unable to boot or no program was running at all. I started to speculate what could be causing the problem, so the first thing I thought was to try to dissipate any possible charge stored in the caps involved with the encryption circuit. Tested the board, but no change. Second, I checked the PRG roms for integrity and everything was OK. Third, I placed the Razoola PCB tester ROM and from this point on I knew it was really something more tricky/serious than I was expecting as the system didn't boot anyway. I grabbed my logic tester once again and focused on the PROGRAM section. It could be something related to the Q-SOUND program also (ROMS 1 and 2), since the system won't boot if the Q-SOUND can't be initialized, EVEN USING RAZOOLA PCB TESTER. But this was discarded since there was sufficient activity in the address pins or the expected logic state at least in pins such as OE(-) and CE(-) and WR(-), the control pins. So, probing around the main program area I quickly found the OE(-) pin of every program ROM just floating. All OE(-) pins of all program ROMs are tied together and I ended up tracing it back to the pin 12 of a 74LS245 three-state output bus transceiver, at region 10F, that simply controls the bus for the main program. Pin 1 of this IC defines the direction for the bus, depending on its logic state. In our case, it was well defined since it was tied to VCC. If it was HIGH, then the bus direction was set from A to B. Pin 12 (B7) was and output, a floating output. Pin 19 is the G(-) or CE(-) and it was LOW as it should. Pin 8 is the A7 and it is the input for the B7 output, and its state was HIGH. Since the chip was enabled, this pin should be at least HIGH (as everything should be pulsing), not floating. Made a small surgery by replacing this suspicious 74LS245 for a brand new one and............... Problem solved! One more CPS 2 B board brought back from the dead. The board booted normally and I played through the game from start to finish with no issues. I rank this an easy fix. :D
  2. Do you know one thing? I started trying to repair one of them, since they had the same problem. I replaced too the original Z80 for a new Z80A on the Captain Commando pcb, but on the SFIICE I left the original Z80 there and replaced only the oscillator. Only the oscillator replacement fixed both boards. Friend, when I say it was bizarre to get two boards the same day with the saaaaame problem, and from different sellers, you don't believe me.... Take a look at the xtal you removed from your Final Fight, if you still have it. I'm now looking at both faulty ones I've removed and they have the following writing on them: VX-8111 / 3.579545 MHz(obvious) / JVC / 1E (one), 1F (the other one). The two bad osc are of the same brand. 1E and 1F sounds like date of manufacturing or something the like. Pay attention if you are having sound problems with CPS1 A boards. Before replacing Z80, rams or something with more legs, look at the osc. If it's exactly the same part I described, take the time to analyse if its pulses fail. In fact, an active osc is nothing more than a reduced version of a real size osc composed of a Xtal + 2 gates of an inverter + 2 or 3 small ceramic caps + 2 or 3 resistors inside a shielded cage. An intermittent fail in one of the gates of the inverter is probably the cause of the pulse getting stuck sometimes. It is sufficient to crash the audio subsystem.
  3. Problem solved! Last Sunday I've fixed both boards. The first suspect for me it was the audio circuit clock, but I analyzed it in two situations only: When the audio was ON and when the audio was already OFF. This small mistake led me to replace the Z80, the audio ram and to program new GALS as substitutes for the SOU1 and IOA1 PALS unnecessarily. Despite the fact they could be related to the fault, nothing changed after replacements. I tried to fix first the Captain Commando, since the SFIICE had the same issue. So I imagined a third situation: The possibility of a clock signal random inconsistency... I placed the logic probe's tip on the 3,579545 MHz oscillator's clock pin and then powered up the board. I kept holding the tip on that pin for about 7 long minutes watching healthy pulsing lights and listening to the game's normal sounds until...... All sounds gone away and the clock pin got stuck LOW for about one second long and started pulsing again, but it was too late. Everyone involved in the sound system got lost and didn't know what to do next... Replaced the 3,57MHz oscillator and the sound issue was gone. I did the same with the Street Fighter IICE A board that came with the same issue without even testing and ...... Problem solved too. Double victory!
  4. Updates.... I took the time to analyze logic states around the audio section and at the moment I got the following: When audio is on, there are lots of activity in every pin possible, with good healthy pulses. But when audio goes out, all address lines and I/O pins become floating. No activity at all. Same happens with the YM2151 and audio ram (2015, 2018, 6116...). I started by removing the Z80 and installing a socket in place. I seated other Z80, this time the correct one, a Z80A, but no success. After a while the issue happened the same way. Later, I decided to replace the audio ram don't believing this could make any difference, but I tried. installed a 6116 in place of the original 2015 and ...... no change. Analyzed the pins of eprom 9 (audio PRG) and found that when the audio goes out, address inputs get stuck high for the most part. A5, A4 and A3 get stuck low (pins 5,6 and 7). All outputs get FLOATING and the control pins, CE(-) and OE(-) get stuck high. This explains why we have no outputs. According to the schematics, the eprom 9 CE pin comes from the pin 12 from SOU1 PAL through CN1, pin B4. So I programmed a GAL16V8 with the corresponding .JED, but no change. Z80 reset line is OK. Clock line is OK. The biggest hassle is that I'm working on the smaller pcb that's full of unreachable smds:x I would really like to replace one 74ls138 as it does the CS(-) functions for both OKI and YM2151 chips.......... And I have no sound at all... And a pair of 74ls374 also controlled by the above mentioned 74ls138, that seem to receive pulses from the 68k and then sends them to the sound eprom's address inputs. They are fed by a clock pulse that comes from another PAL, the IOA1 that passes through a 74ls32 at region 5B, with outputs at pins 6 (sound 1B) and 8(sound 0B). I imagine that without this clock pulse, no audio at all! Also a 74sl74 that seems to control the SOU1 pal and the select pin from the OKI, but it seems a bit out of question... These are the suspects at the time. The reason I didn't probe them yet is due to the fact they aren't accessible when the b board is connected since they are smds. Comment people! Our beloved hardware, marvelous piece of past is getting older and older. I'm tired of seeing dead or partially working A boards with almost no hope of fixing. I ask it with passion. Let's try to fix them! Kaisersigma. In case someone is interested, I have the schematics for the CPS1 A and B boards.
  5. Hi. No, my super gun does have a switch to alternate between mono and stereo, but it always stay in the mono position. When not using a consolized unit for playing MVS games, I use a MV-1FZ with the super gun which outputs mono audio.
  6. Hiiiii, Womble. I'll reproduce the fault again (I know this will happen again :x) and will check if it gets really hot. Very noisy PS? Would I need an oscilloscope to check it? The JAMMA connector is pretty clean and shiny. I use all my PCBs with my super gun and to generate the audio line level signal I use audio transformers with the right calculations not to stress or damage power amps. It only happens with these two last boards I've just bought. The audio of all my other pcbs just work flawlessly, including 9 other CPS1 pcbs, between dash and 10Mhz versions. But, how can I check my PSU without an oscilloscope? I use an arcade PSU Eago, 15A in the 5 volts line, 2,5A on the 12V and 1A on the -5V. Power is regulated on 5.05V. interference? Would be some sort of signal flowing through the 12V in case of any issue with the amp, or in the 5V line if it's a logic issue? What worries me is that these kinds of issues are becoming more common. It's the first time it happens to me, in my case in a double dose, but I heard the exact story from others. Thanks a lot... - - - Updated - - -
  7. Yes, I've been thinking about it with Captain Commando but the SFIICE makes strange noises after the audio goes out. It's not a noise of something trying to make contact again when the audio would be still playing but interrupted of reaching the amp, but sounds more like quick pieces of garbled audio, as if the........ YM synthesizer(?) didn't know exactly what to play.
  8. Hi guys. :) In the same week I received two pcbs, Street Fighter II CE and Captain Commando, both originals and with the same bizarre issue. The games work normally but suddenly ALL the audio disappears, but if I turn them off then on again the audio comes back until it disappears again. I left the SFIICE turned on for a fair amount of time and the issue mysteriously stopped happening, but it still happens with Captain Commando and I'm left with the question: WHAT COULD BE HAPPENING TO CAUSE SUCH AN ISSUE???????????? 6116 ram, Z80, ym synth, 3014B??? The audio PRG is ok so what's happening??? Any help would be appreciated and I know others do have these kinds of issues. :(
  9. Today I've built a circuit that solved the audio problems with one of my toki pcbs. I had only samples but no FM. I tried to feed the mixer channel that receives the sampled audio with the FM from the Y3014b but the result was volume imbalance. The samples were overshadowed by the much louder FM. In some way the original epoxy mixer could balance things, tie the sources together, buffer them and send to the Power amp. Analysing the Y3014B, any dual opamp and the LA4460N Power Amp, I could trace what one wants from the other. So... the tests began. Y3014B wants to know the opamp gain needs/limits to work right. So it sends the audio at the right level. If you do not do this its output will be distorted. I used a LM358N from a really faulty board and everybody knows it has bad reputation in many ways. It is a bit noisy and tends to distort high frequencies. The 222 ceramic cap (2200 pF) in the FM line tied to ground reduces a bit the distorted high freq impression without much losses and can be omitted in case you use a better opamp. The 20k in the FM line is there to adjust/reduce the Y3014B higher gain in relation to the 2,2k in the samples line. This way all sources will be at the same level. No need to have -5V in this board to have sound anymore since I used a single power supply opamp. I suggest any other single power supply dual opamp to replace the LM358 due to audio fidelity, but the circuit won't change. you'll have to remove the original sip epoxi and connect in its place the pins shown in the diagram. In the board you'll see pin 1 to 20 from right to left.
  10. Hi Jimmy. It's been a while since I tried the freezing spray without any discoveries or clues. Today I discovered one thing about this Toki PCB I would like to share with you all. First of all, My setup consists of a supergun that powers the board using an arcade PS that can deliver up to 15A through well dimensioned wires. So no power problem! And............... I use the Tototek adapter to connect PSX pads. I have some of these pads here with me and I picked up one of them that was damaged and then refurbished and I decided to use it specially to play fighting games and it ended up being the main pad for my supergun. Newer PSX pads have a plastic limiter in the form of a X that sits on the D-pad not allowing more than 2 directions (diagonals) pressed at the same time. I replaced my pad's broken limiter with a piece of hard rubber just to be able to use the D-pad again. The problem is that in the heat of the moment, if you press the D-pad a bit more you'll end up pressing more than 2 directions. Maybe 3 or the 4 directions pressed at the same time. TOKI seems to not like it. I can reproduce the freezing issue now when I please. I made it 5 times in a row in the first stage's first screen by pressing these many buttons. This now explains why the PCB will not crash when left powered on for a long while alone without someone playing it. This discards the overheating possibility too. I tried the game with another PSX pad (now perfect) and it played fine. I never knew in my entire gaming life of a game that crashed by receiving many inputs. I wonder what happens internally to cause such an annoying problem. I have another perfect TOKI pcb here and I'll test later to see if the same issues happen by receiving these many inputs. Until next time.
  11. Thanks a lot, Jimmy. I'll search for the freezing spray and try the tip. I've dumped all roms, sent to romident and all of them are fine. No bad roms.
  12. I measured the voltage directly in the chips pins near the jamma connector and I got 4.98V. In distant places 4.88 ~ 4.90V. I powered on the PCB and left it running alone (the presetation and demonstration screens) for about 2 hours and went to do other things. When I returned, the PCB was still running fine. So I decided to play once again. I played for about 15 minutes from beginning to Bashtar (stage 5 boss) and then the game froze again. It seems the freezing issue happens only if there's someone actually playing the game. Any clues? Do you think this alone discards a power supply issue?
  13. So I reinforced the +5V and GND soldering in the power supply to enlarge the conductive surface and corrected a mistake I made when I removed the PS metallic shell. Basically the shell shortens some points and all I made was join them again. Now I'm reading 5V instead of 4.86 (when connected to Toki) I'm with a sense of dread because when I read the output voltages before this correction, it was like the PS gone mad! In the +5V line, the voltmeter was going up and down from 0,54V to 4,7V like mad, almost in fixed intervals. Same for the 12V line. The -5V line was normal. I found strange when I inverted the voltmeter tips. I was reading now fixed -5V in the +5V line without the mentioned oscillation and -12V in the 12V line. Even with these fixes, the PCB was still freezing. Then I removed every socketed chip in the board, cleaned the pins plus sockets and reseated them all. I always use antistatic wrist strap to perform such a task. Tried again and..... PCB STILL FREEZING!!!! Below is a video I made that shows what actually happens. https://www.youtube.com/watch?v=YJXRLywWjfs Any help would be welcome. My wife told me to forget about it for a while, but I really want to see this thing working as it should....
  14. Oh, Womble... I measured the 5V rail before a coffee break and I've seen it falls to 4.83V ~ 4.86V. I'm worried because if Im having this voltage in the JAMMA connector, maybe I'm having even less in farther regions of the PCB. Maybe all the logic is being underfed.. This is the supergun I built myself with audio attenuation, odd sync pulse fixer, composite and component video out, and a PC power supply. To the right is the power supply. I used four wires for +5V and 4 wires for GND to ensure sufficient power. When I was in a test phase, I did not experience any trouble maybe because I used all +5V wires coming from the power supply (at least 8 red wires) tied together. Before I install the PS in the supergun case I tried to make it clean by removing all unused wires. I'll make some tests. Thank you very much... Alex
  15. Thanks, Matthew! I'm feeling at home...:)
  16. Toki pcb randomly freezing!!!!!!! Hello to all of you arcade dreamers like me. I have a Toki PCB that boots ok, plays ok, sounds ok, but lately something is REALLY BUGGING ME. I'm experiencing RANDOM FREEZES of a strange kind. The game suddenly stops! The screen gets static, the music continues playing normally and the game accepts more coins if I press the coin button (I can even hear the insert coin effect). I removed and resoldered the 20 Mhz oscillator and nothing changed. By inspecting, I can say the solder joints are OK. All I can say is I love this game and I would really be grateful for any help. Please help-me!!!!!:(
  • Create New...