Jump to content
Due to a large amount of spamers, accounts will now have to be approved by the Admins so please be patient. ×
IGNORED

Sega Zaxxon Repair Log


Recommended Posts

Some more stuff on the old to-do pile was this pair of Sega Zaxxon PCBs.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_4553.jpg

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_4554.jpg

 

One of them was the board I had fixed the sound fault on ages ago, but it had developed a graphics fault soon afterwards and gone back on the pile. I suspected it was my fault as I had left some foam in between the boards when I put it back together, and the graphics board got stinking hot as a result. I had also pulled it out about a year ago to try to help out a KLOVer with a similar sound fault, and found it to be totally dead at that point, giving a blank screen and no signs of life.

 

I sometime wonder if stuff this old just wants to be left alone to die.

 

The second board I got in a large batch from somewhere years ago, and it had never worked, giving a screen of rubbish on power up. When I got it was was labelled "Super Spy??", god knows why as every single EPROM label on it clearly says ZAXXON.

 

Anyway - seeing as I figured I had cooked something one the 1st board, and probably something hard to get like the strange RAM that was the source of all that heat, I started on the "never worked" board hoping for an easy win.

 

On board up it gave this.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3928.jpg

 

These boards are a 3 board stack, two main boards, a GFX PCB at the bottom, CPU PCB in the middle and a small audio generator board on the very top. The audio is a completely separate system on these so I started by taking that board off so I could get to everything on the CPU board.

 

With the usual tests on any board completed ( check the voltages, reseat everything socketed while checking for bent pins, and remove and reseat the ribbon cables) I started with the CPU, a very swanky looking Sega badged Z80...

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3929.jpg

 

....and the fault was pretty clear as the /RESET line on pin 26 was constantly low, so the board was stuck in reset.

 

I couldn't find the exact schematics for this 3 board set, the more common pcb is a more rectangular 2 PCB stack, but I could get the gist of the reset system, and the fault was pretty obvious when I went poking around in that area.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3927crop.jpg

 

The PCB has three 10uF 16v electrolytic capacitors on the board and all three had leaked and were very crusty. One is involved in the Test switch signal, and one is in the Reset system, its purpose is to cause a small delay on power on to hold the board in reset state until the power supplies have stabilised. With such a rotten cap it never charged to the point it took the handbrake off.

 

Three capacitors out...

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3926.jpg

 

and three replacements in...

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3931.jpg

 

and that's one board fixed!

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3932.jpg

 

PCB board faults are almost never due to bad capacitors (with the exception of some sound faults that are down to amplifiers issues, the plague that made bad capacitors infamous didn't kick off until the late 1990s and those were usually caps in power supplies or DC to DC converters that get driven hard,

 

On to the second board, and with the same degree of crustiness present I just replaced the three 10uF capacitors there too...

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3936.jpg

 

...and that woke it back up again, revealing the graphics fault I thought I had caused a few years back.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3940a.jpg

 

The game runs fine but the landscape underneath the ship is made up of the wrong tiles, and when the ship gets out into space the planet backdrop is made up of incorrect graphics elements too.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3951.jpg

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3946.jpg

 

As I had a fully working board-set I swapped the good GFX board from the other set onto this CPU board, expecting the fault to be cleared, but it was still there, so the issue was actually on my CPU board and not related to the hard to get RAM chips that I suspected I had cooked. Completing the test by confirming the graphics board worked fine with the other CPU board meant I had halved the area I needed to search in, which is always handy.

 

Poking around with the scope I could narrow the fault to the 3 ROMs labelled 4, 5 and 6.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3970.jpg

 

Shorting data lines on those cause the faulty sections of the screen to go haywire. I would ideally have read the EPROMs at this stage, but they were 2564s rather than the much more common 2764 type. I do have an old 1980s Data IO 29B reader which can read these but its a bit of a mission to get it set up with a DOS PC and monitor, so I just assumed they were OK for the time being.

 

I ran the scope over the shared address bus to see if they were being correctly driven, everything looked busy and healthy except pin 6, which is A4, having a pin doing nothing in the middle of the address bus makes no sense from a design perspective. This would look healthy for some of the time but when the graphics fault was on the screen it would go totally silent and sit low.

 

This line tracked back to a 74LS374 octal latch at U76 (U114 on the schematic for the alternate board layout schematics I found) on pin 5, which is the output pin corresponding to input pin 4 which was behaving the same way, silent when the fault was evident, and busy when it wasn't, so the issue was further upstream and not the latch itself.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3969.jpg

 

This input pin tracks back to pin 12 on ROMs 7 and 8. It's unusual for ROMs to directly drive ROMs, but it explains the fault as I suspect this is how the background scrolling is achieved by streaming data out of the ROMs that then selects the tile data in the second set of ROMs.

 

ROMs 7 and 8 operate together, but only one is active at any given time, their /CE (chip enable) pins being controlled by upstream logic to ensure they time-share the bus correctly. When ROM 8 was active there was a healthy signal from its pin 12 present on the latch input, and therefore the output which was driving A4 on ROMs 4, 5 and 6 was also OK. When ROM 7 was active there was no signal into the latch and the fault occurred, so the fault was right at ROM 7, and after pulling it from the socket it was obvious what was going on.

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3965.jpg

 

The issue was down to badly tarnished pins on ROM 7, pins 11, 12 and 13 were far more oxidised than the others, with pin 12 being the worst of the lot. After a lot of polishing with a glass fibre pen the connection to the socket was restored and the fault was gone!

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3953.jpg

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3957.jpg

 

http://i40.photobucket.com/albums/e230/Womble76/Zaxxons/IMG_3961.jpg

 

Three simple fixes, and two boards alive and kicking again.

Edited by Womble
Link to comment
Share on other sites

Thanks for sharing. We're the other ROMs on that board as tarnished as ROM 7?

 

Nowhere near as bad, only the end four pins on that one IC. Usually a pull and a reseat scratches through the oxide layers but these were black and crusty.

Link to comment
Share on other sites

 

PCB board faults are almost never due to bad capacitors (with the exception of some sound faults that are down to amplifiers issues, the plague that made bad capacitors infamous didn't kick off until the late 1990s and those were usually caps in power supplies or DC to DC converters that get driven hard

 

 

I once had a Xain'd Sleena bootleg with a strange fault. When you got to the planet select screen, there was no ship to move, or planets to select. It continued the music to the end of the track and then would halt on this screen. I chased my own tail for a long time trying to find the cause but gave up. It went to the to 'do pile' for months, until one day I acquired another working boot, so I set out to compare it. I can't remember exactly what it was, but I think I found one data line on the working CPU board on the second 68000 tied high, where it was active on the non working board (or something along those lines). I traced that line back through some circuitry, and then to a 4.7uf cap positioned alone at the edge of the board. The cap on the non working board had one of the legs popped off. With very little confidence (there's no way a 4.7uf cap could cause an issue like this) I replaced it and BAM, the board worked perfectly after that. This is the only board I've worked on with something as bizarre.

Edited by Holy-SNES
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...