RE: Tempest CPU problems...

From: Rosenzweig, Joel B <joel.b.rosenzweig_at_intel.com>
Date: Mon Apr 22 2002 - 12:59:18 EDT

I'd consider replacing the remainder of the RAM. It just feels like one of
those errors. Another possibility is an intermittent connection on mathbox
to the main board, either at the pins or the cable. I've seen really weird
stuff happen to Tempest when that cable is defective.

Joel-

-----Original Message-----
From: TomWiz [mailto:twisnion@enteract.com]
Sent: Monday, April 22, 2002 12:47 PM
To: Vectorlist
Subject: VECTOR: Tempest CPU problems...

Guys,
  I'm looking for some opinions on repair paths for a Tempest
boardset
that is driving me nuts. Here's the history.

Boardset was in cab that was exposed to water. Board is clean,
but dirty.

Initial power up, boardset dead.
Several Roms had oxidized legs, replace roms, board now boots to
ram errors.
All roms good, supply good.

After fighting with the RAM errors (5 Bad Rams total, 4 Vector
Rams, 1 Pgm Ram)
game now boots clean but no XY Output.

Digging into XY Output stages and found +15 regular went nuclear
at one time,
working from the bottom up, ended up replacing every (ouch)
component that
was on the +15 line. I now have clean vector output on scope,
time for the
monitor connection.

Mathbox tests reveal two bad pokey's, replaced them. Now no
errors on
any test (Slam etc. Test screens are perfect.

BUT, during intro screens I have some oddities.

A. High score display is showing.
     1. Jim
     2. joe
     3. tom
     4+5.&^%% (This line has score 4 and 5 superimposed on each
other.)
     blank line
     6. bob
     7. you

B. Intro screen where Tempest grows from center has two Tempest
logo's that
grow in sync (kinda cool) until approx center of screen, then you
get one as
normal. As they grow, they correctly have the color 'sub logos'.

C. During game play, the Green Spikes will 'Sometimes' start off
at the wrong
spot and move at odd angles. An occasional object will jump out
of the tube
and then back in. You can still shoot the items like there were
in the tube,
so that leaves me to believe that Ram has the correct object
positions reflected.

A known good mathbox was swapped in, no change.

I've gone thru with a Huntron Curve tracer the vector pgm
counter, buffers
and a few of the VSM chips, and nothing was out of line. But I
didn't expect
I would find anything with the tracer with this type of failure.

I've checked the archives, and there was one other person with
the
similar problems that posted, but there were no public replies. I
know I've
seen this before on the list...

So, with all of that said, can any experts to the inner workings
of the
CPU point me to an area that might be worth investigating. I hate
using
the 'Brute force method' of just replacing stuff without a
direction...

Thanks
TomW

---------------------------------------------------------------------------
** To UNSUBSCRIBE from vectorlist, send a message with "UNSUBSCRIBE" in the
** message body to vectorlist-request@synthcom.com. Please direct other
** questions, comments, or problems to neil@synthcom.com.
---------------------------------------------------------------------------
** To UNSUBSCRIBE from vectorlist, send a message with "UNSUBSCRIBE" in the
** message body to vectorlist-request@synthcom.com. Please direct other
** questions, comments, or problems to neil@synthcom.com.
Received on Mon Apr 22 10:08:04 2002

This archive was generated by hypermail 2.1.8 : Fri Aug 01 2003 - 00:34:00 EDT