STAR*TECH Journal
Case Study #903/ Atari
GAUNTLET No Picture
By
Sam Cross
Godwin Distributing Co.
North Little Rock, Arkansas
GP.ME: Atari GAUNI'LE"I'
TROUBLE: No Picture
1) Tried to put a credit on and ·start game
to listen for game sounds, but there were
none.
2) Took off the back door and was going to
measure for +SVOC on board, but Atari has
a +SVOC LED indicator on the board and it
was lighted so I figured there was +SVOC
at the board.
3) Since most boards I encounter will
display a picture with a +SVOC power
supply, I then thought the monitor might
be the problem so r ...
4) Took the upper door out and saw the
filaments in the neck of the picture tube
were lighted, so I figured the monitor was
good, but as a second check, I turned up
the brightness to see if I could see a
raster pattern, and I was able to see one
-so I more or less guessed the monitor was
working.
5) Thinking that the logic board was at
fault, I checked for crystal activity with
II'!Y logic probe on the board. There was
activity. Still, I reasoned that since
most boards will display some kind of
picture with +SVOC only, I decided to see
if there were any other voltages missing.
6) I got the schematics out and saw there
was +15 volts supplied to the main logic.
I checked at the CPU power plug and found
no +lSVOC.
7) I found by the schematics that it came
from the audio board, and then found the
AC fuses blown in the bottom of the
cabinet that supply the AC power to the
audio module. I replaced them and they
blew i.rcurediately. Figuring that when fuses
May 1987
[ 5)
blow immediately, it is usually because
one of the rectifying diodes was shorted,
r ...
8) Checked the diodes with my ohrnneter and
found that they were good, also I saw
these four diodes were arranged in a
bridge rectifier configuration, so I
checked for resistance across the positive
and negative terminals of these diodes and
found there was only 8.2 ohms resistance.
Since from the schematic I saw that these
diodes supplied the +15 VOC and -lSVOC
voltages, this means there is normally 30
volts at the+ and - terminals, and with
only 8.2 ohms resistance across the bridge
that means that over 3 arrps would flow. I
had the audio module on the bench so I
figured that I would check for a shorted
corrp)nent on the audio module.
9) Seeing two Tm 2030 arrps were used on
that board, I unsoldered one of the power
supply legs on each amp and found only 8.2
ohms across one of them. This one I
figured was shorted.
10) The 'IDA 2030 looks just like the more
co:rrm:m 'IDA 2002, yet the '2030 operates
from a split supply and the '2002 operates
on +lSV alone.
11) Not having a '!'DA 2030 on hand, I had
to modify the circuit to use the 'IDA 2002,
and it wasn't that difficult.
12) When I replaced the anp and the fuses,
the game started working.
13) Hindsight is 20/20. Next time if I
ever encounter this problem, I will know
what it probably is, but the first time is
the hardest.
Murphy's Law #903
Indecision is the basis for flexibility.