Sega G-80: Troubleshooting cpu boards

Zitt

Member
Joined
Sep 6, 2008
Messages
767
Reaction score
15
Location
Austin, Texas
Now that I received the donor captain's chair; I've begun checking out the board set in lue of a vector monitor.

I've hooked up an oscilloscope in X-Y mode to the X & Y outputs on the video cable - in a bench configuration. By Bench config; I have a 120VAC plugged into the transformer at 1&3 and measured the voltages 5, -5, 12, -12 on the PSU connector. All measured in range.

Other than X/Y (J3); no other cables are plugged in.

When the boardset is first powered up ( I removed the sound boards) the CPU board LED flashes once briefly. The Oscope shows a 45 degree line.

Pushing the test button; the led flashes intermittenly several times then goes out w/ no change to oscope.

I've run across a couple of simple troubleshooting tips - are there any more detailed simular to the pinball repair archives?
 
I got to looking at the cpu board.
It appears that someone has clipped off Y1 (8MHz) and the two 330ohm resistors.
I have no clue why someone would clip off componets like that.

I looked all over my house and no 8MHz crystals ... even checked "parts" motherboards.

<sigh> At least it explains why the CPU LED was "odd".
 
I too have a captains chair that is not working either.

..at least you are getting a LED that lights sometimes. I am not getting a lit LED at all.
Like you, I'm getting +5 -5 +12 -12 to the harness off the PSU.

Let us know if you have any luck when replacing the crystal. I don't think that I'm having the same problem with my CPU board tho.

I'm stumpped for now...

-Muel
 
You won't get much of anything without that crystal- that's for the main clock.

If you're having trouble after the clock is running, try replacing the Z80 or those 2114 RAMs. They're socketed, so that's a quickie.

K
 
You won't get much of anything without that crystal- that's for the main clock.

If you're having trouble after the clock is running, try replacing the Z80 or those 2114 RAMs. They're socketed, so that's a quickie.

K

From the schmatics, the CPU actually gets its 4MHz clock from the XY Timing board (The timing board grounds 17-ExtClkSel, the CPU gates over to T-ExtClk). The onboard xtal is for troublshooting only and should not be needed...
 
From the schmatics, the CPU actually gets its 4MHz clock from the XY Timing board (The timing board grounds 17-ExtClkSel, the CPU gates over to T-ExtClk). The onboard xtal is for troublshooting only and should not be needed...


Oh- I see-
The Z80 on the CPU board is clocked by either the 8MHz on the CPU board or the 4MHz from the timing board as determined by the pin 17.

That's clever. So, I guess all of the Sega vector games are driven by the XY timing board, right? I wonder if the Sega raster G80 games make use of that or if they use the CPU crystal.

Thanks for straightening me out. I saw the crystal feeding through some circuitry into the Z80 clock pin and ignored the rest of that...

I see the pin 17 is grounded through the XY board. With it pulled, the CPU will revert back to it's own clock? You couldn't see any output with the XY timing board out, but I wonder if anything would play?
 
From the schmatics, the CPU actually gets its 4MHz clock from the XY Timing board (The timing board grounds 17-ExtClkSel, the CPU gates over to T-ExtClk). The onboard xtal is for troublshooting only and should not be needed...

humm... wish I'd seen this a little while ago. I replaced the clock section on the cpu board and got it to work.

Started a worklog too:
http://forums.arcade-museum.com/showpost.php?p=1505972&postcount=2

I'll have to go back and look at the XY timing board.
 
OK- somebody explain that one to me then.

Is the Z80 on the CPU board getting the clock from the CPU crystal or the XY crystal?

You could tell by looking at pins on U18 on the CPU board. Is pin 11 or pin 8 clocking? The /EXTSEL on pin 17 of the cage should choose one or the other.

If XY Timing is grounding that signal, then why does it matter if the CPU crystal is running?

I see that a clock/2 is generated from the CPU one, but I don't see where that connects to anything.

Nice writeup on the other page. Nice to see the Star Trek up on the scope too :)
 
OK- somebody explain that one to me then.

Is the Z80 on the CPU board getting the clock from the CPU crystal or the XY crystal?

You could tell by looking at pins on U18 on the CPU board. Is pin 11 or pin 8 clocking? The /EXTSEL on pin 17 of the cage should choose one or the other.

If XY Timing is grounding that signal, then why does it matter if the CPU crystal is running?

I see that a clock/2 is generated from the CPU one, but I don't see where that connects to anything.

Nice writeup on the other page. Nice to see the Star Trek up on the scope too :)

Not sure to be honest.
I did notice that I had an intermittant problem with the XY boards. I pulled them and found that it appeared someone had bent one of the pins on the 24pin DIP ribbon innerconnect. For safety; I also resoldered the IC sockets under the innerconnects incase of a cracked solder join. It now comes up reliably.

I'm wondering if that it why my xy board was not generating the Clock for the Z80 CPU.

The problem with your request is that damn cage. Nice to easily swap boards... not nice when you need to measure something in the running system. There isn't any easy way to probe the system while it's running. It appears the CPU board only needs +5V. But it may also be using the 3vac line to "activate" reset on the cpu. As such; I haven't determined if it's possible to run the CPU board completely on the bench.

A quick gander at the schematics - the /extsel pin ground one of the gate lines into a NAND gate. Which in effect prevents it from ever going low.. which means the crystal's CPU logic never oscillates at 8MHz. This allows the extclk to pass through the nor gate tied to the cpu clock - allowing the xy board to deliver the clock.

If they didn't do that in the logic; both clocks would appear at the nor gate making it impossible for the cpu to sync the clock.
 
You are welcome to come to my shop and test your boards in my working ST ur. I also have a few extra sets we can swap around with.

I will be messing with the new Sega Multi kit I got from Vectorlabs over the next few days so all that stuff is out where we can get to it.
Just lmk.
 
I once tried to pull the backplane out of the cage, but that made for a sloppy stack of boards that you have to be careful with... You can plug the boards in any order in the backplane, so the board of interest can go in the top.

I think I finally just resorted to extending the signals of interest using wrapping wire, and terminating on a barrier block so I could measure outside the cage.
 
Back
Top Bottom