Need Ms. PacMan help again...logic probe

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Need Ms. PacMan help again...logic probe

Post by ModRob » Tue Dec 14, 2004 11:29 am

Well, I've been working with the logic probe, and digging myself in deeper.
I think I found that the watchdog was "barking", in that pin 15 on chip 9c was chirping. After some more poking around, I replaced both 8c and 9c. Now I'm getting active states and pulsing. Pin 6 on the Z80 is a fast pulsing, but pin 26 is now in a constant LOW state. I tried to use my schematic and backtrack, and somehow thought that there may be trouble in the chips at 2r, 2s, 3r, and 3s (I believe "counters"). According to one site, I should have had pulsing on the outputs of all of them at pins 11 thru 14. I had nothing if I remember correctly. I then socketed them and replaced the chips, and now I get steady states on 2r and 2s but no pulsing. On 3r I now get pulsing on all 11 thru 14. But on 3s, I only get pulsing on 14, and steady states on the others. Of course after everything I've done, I plug the board back into the game, and still only get a light green screen to come up; no action, not anything else. NO sounds. Nothing but that blasted green screen.<p>This after six hours this morning. Again, appreciate any direction you can give me...

hassan1980
Posts: 7
Joined: Tue Dec 14, 2004 1:01 am
Location: Syria
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by hassan1980 » Tue Dec 14, 2004 12:22 pm

Hi,<p>Two months aog, I made a logic probe based micro 12c508.<p>if you are interested, let me know<p>Hassan

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by ModRob » Tue Dec 14, 2004 3:41 pm

Oh, and another thing..I replaced the Z80, but all I could find was one labeled "Z80A"...I see on another site where it's a higher--4mhz, compared to the 2.5mhz of the Z80. Does this harm anything?

Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by Enzo » Tue Dec 14, 2004 11:25 pm

A Z80A is fine. Without getting the reset thing right, nothing else will happen. Pin 26 needs to sit high. WHen the watchdog cycled, 9C was doing its job. It was getting no clear pulse, so it kept counting out and triggering the reset. That is its job. The timer clear pulse is generated by the Z80 running the p[rogram.<p>The master clock is upper left, a crystal driven by 8B. SHould be 18MHz at 8B-7. That is then divided by 8A, which puts out 6MHz clocks into the bus. Those are 8B-13, 8B-3, 4F-11. Note that 8B-13 also sends its 6MHz down to clock the dividers 3S, 3R, and 8C right below. Those dividers spit out the H signals. Pins 11-14 on 3S, 3R should be active. Also 8C-5 is active with the last H signal, 1H. All those need to be active all the time.<p>Note that signal 1H is also the clock for the Z80, inverted into pin 6 of the Z80.<p>Just to the right of those is an identical circuit producing the V signals. Unlike the H chain, this V chain does not divide off the 6MHz, it is clocked by the H SYNC generated below in the group centered around 3N. That H SYNC is formed by combining various H signals logically. SInce all the Hs are always active, H SYNC will be too. H SYNC is divided into all the V signals by 2R, 2S, and 5M. All of those should be active.<p>Below the 3Ns we see 5M - the other half of the 5M we mentioned a second ago. 5M spits out V BLANK. So far all the stuff we have seen shsould work all the time. Pull the Z80 and they still should happen. They are your various clock and timig signals. Everything else int he game runs off them somehow.<p>Collectively, all these are the timing circuits. Unless they are running, nothing else matters.<p>We need to see good clock at pin 6 of Z80. And we need to get the reset line HI.<p>9C is the watchdog. It is just a counter constantly clocked by V BLANK coming from 5M to its left. pin 15 is marked TC - terminal count - it is the "carry" pulse from the counter to trigger the next counter should there be one. Like an odometer sends out a signal to the tens dial every time the ones dial gets to nine. 9C sits there endlesly counting, and every time it gets to 16 it sends a pulse out the carry and starts over. It will do this unless something stops it. That carry is the reset line. Invert that through 7L and it goes right to the Z80. <p>Pin 1 of 9C is the clear. Pulse that and the counter clears and starts over. The prrogram normally sends a pulse repeatedly to that clear before the counter can time out and send a reset. SO it counts all the time but never is allowed to finish. That way the CPU never gets reset. If the program stops running, after a second, the 9C will time out and send a reset. Hopefully it restarts the computer and all is well. You are not so lucky. Flipping DIP switch 8 ON should freeze the video and halt the program. <p>If you ground 9C-15 or pull the chip, then the reset on the Z80 should rise and the thing is free to do whatever it can. Maybe something maybe nothing.<p>Have we flipped the test switch inside the coin door over to the TEST position to see what happens?<p>The ROMs in row 6 are the program ROMs, and those in row 5 are the character ROMs If the program is running and you pull the row 5 ROMs, the Pac Man becomes just a square moving about rather than the cute graphic.<p>Omce the timing is running and the CPU is clocked and not held in reset, we can try to determine why things are not progressing. The CPU looks to the row 6 ROMs for its program. Even that doesn't get far. The CPU has a bunch of address lines and a few chip select lines to control the ROMs. and a data buss to read what they have to say. But for anything else to happen, the CPU has to communicate with the rest of the world. And to do that it need a good custom chip or a good rider board at 6D - the Z80 Sync Bus COntroller. It coordinates the CPU data buss with the game data buss. Sorta like your lungs coordinate the oxygen in the air with the oxygen in your blood. Sorta. 6D is next to the Z80.<p>Now look over at the other custom IC or rider board across the way from the Z80 at 5S. That is the VRAM addressor. The underside of the chip is its imput. Note they are all V and H signals and so should be present always. Other than FLIP ther are no control inputs to the c hip, so I expect the outputs to be alwaysactive too. This rider is a lot easier to check than the other one.<p>Logic chips rarely survive reverse power connection.

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by philba » Wed Dec 15, 2004 5:15 pm

I assume you don't have a scope. That would really help in debugging this problem. Very quickly you can see if the z80's getting a clock, if the h sync and v-sync are working, etc. Maybe you could borrow one.<p>You might want to look for solder fractures in the board. I've had non-functioning equipment come alive after I hit the solder joints with a hot iron.

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by ModRob » Thu Dec 16, 2004 12:59 am

Enzo..again, thanks for so much info. It's a bit over-my-head, but I'm trying to get a grip on it.
I've learned quite a bit so far with your help...<p>And about the scope, I thought that might be a help. I do have a fellow at work that works in our RR signals department--if I can catch up to him, I can get him to scope it out for me. I may have to end up having him help me even more than that, with the luck I'm having..haha...

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Need Ms. PacMan help again...logic probe

Post by ModRob » Thu Dec 16, 2004 1:09 am

Oh and one more thing Enzo...I have flipped the test switch inside the coin door, and nothing at all happens. Of course it did work fine before I started on my "work"...haha

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 31 guests