TECH: Williams ROM board happenings ?! Observations !

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Hello All,

I've been working on some Williams boards and have come across
something interesting that I've not seen before.

Scenario:
Working Robotron ROM board ( call it 'Robo A' ) which is 100% completely
tested known bombproof. Working CPU board, same deal this is flawless

These boards were mounted not to an original metal base but on a wooden
board, the wiring harness is decent 18/20Ga everywhere. Voltages measured
at all the boards are a solid 5.1v. That's both the ROM and CPU board.

So the power up self test passes, text is displayed ok ..BUT.. the
Roboton 2084 title screen has corruptions in the 'W' logos and on the
'2084', also during game-play lots of bits of grunts get left on screen. Sounded
like a faulty SC1, so I swapped it and it was all working fine...
However testing the same SC1 in a different ROM Board which was mounted
to the metal plate it worked FINE !!

Trying the 'Robo A' ROM board with the same problematic SC1 with the new
boardset attached to the metal plate all worked fine (eh??)

The SC1 sockets are all good the board interconnects had already been
replaced with new gold plated ones, and two different PSU's were used.

Then putting the SC1 back in the 'Robo A' ROM board and using it on
the original set up AND connecting the grounds directly together
between 'Robo A' ROM and the CPU board everything is working 100% !
No corruptions and the game plays fine.

Next test... I tried a Sinistar ROM board on the same wooden panel mounting
with mr.problem SC1 and the results were the Sini would work fine except
for RANDOMLY adding credits (!!) Grounding the ROM and CPU board everything
worked just fine again.

Finally removed the second CPU board from the metal sheet, put it on the
wood panel and the same results. No ground, random credits, ground totally
stable (!!)

Consistently without additional grounding the sinistar added credits during
the attract screen, or crashed, and the Robotron corrupted graphics.

Yet with different SC1's both sini & robo would work fine ! I've tested
about half a dozen different SC1's, with two exhibiting this behavior and
four not caring about the additional grounding.

So basically it would appear that if the ROM board and CPU board are not
grounded together some SC1's do not function correctly.

Anyone else experience this ? Anyone care to comment on this behavior ?

And congratulations on reading this far...

Cheers
James

------------------------------
JrokLand http://www.jrok.com
------------------------------

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

I've not had personal experience with Williams video games yet (been
too busy with soooo many other projects to get to them!).

I was well aware that Williams boards have to be grounded to the metal
backplate though since they use it as additional grounding. It's the
same with all the older pins (Sys 7 and such). They WILL run without
being properly screwed to the backplate, but it's not how they were
designed, so it's asking for trouble.

With the pins, it's doubly asking for trouble mind. The amount of times
I've shorted something momentarily because the boards weren't screwed
to the backplane and they moved while disconnecting a connector with
the power on! )

Martin.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Only thing I can say is if there are two different grounds, then the
voltages will be different of course. I was running a PCB on it's own wiring
harness and my sync seperator for my monitor on a different power supply.
Video wouldn't sync. Tying the grounds together fixed it.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

jrok <jrokweb@san.rr.com> wrote:
: Hello All,
: So basically it would appear that if the ROM board and CPU board are not
: grounded together some SC1's do not function correctly.
:
: Anyone else experience this ? Anyone care to comment on this behavior ?

That makes perfect sense.

The current draw on the CPU board is a lot higher than the current draw on
the rom board. Even with the relatively small resistance of the ground
wires will cause resistive drops and cause each board's ground to float
above the power supply ground, and the difference in ground levels eats
into the digital IO voltage margins.

--
Mark Spaeth mspaeth@mtl.mit.edu
50 Vassar St., #38.265 mspaeth@mit.edu
Cambridge, MA 02139
(617) 452-2354 http://rgvac.978.org/~mspaeth

 

[Vaxx]

Archived from groups: rec.games.video.arcade.collecting (More info?)

try early irem stuff. They pass the +5v and the GND between pcb's in stacks
through brass standoffs. Make it hard when you forget to jump the power over
to the other pcb when you have the stack apart.

"Martin White" <guddler@gmail.com> wrote in message
news:1126657548.964307.162180@z14g2000cwz.googlegroups.com...
> I've not had personal experience with Williams video games yet (been
> too busy with soooo many other projects to get to them!).
>
> I was well aware that Williams boards have to be grounded to the metal
> backplate though since they use it as additional grounding. It's the
> same with all the older pins (Sys 7 and such). They WILL run without
> being properly screwed to the backplate, but it's not how they were
> designed, so it's asking for trouble.
>
> With the pins, it's doubly asking for trouble mind. The amount of times
> I've shorted something momentarily because the boards weren't screwed
> to the backplane and they moved while disconnecting a connector with
> the power on! )
>
> Martin.
>