Re: Few Space Duel/Gravitar questions.

From: andre <livnfree_can_at_yahoo.com>
Date: Tue May 06 2008 - 09:23:15 EDT

Cool.. I'll print this off and try some things. Have a
logic probe but still not very sure on what to do with
it except check if things pulse or not...I do know
high/low on it but not when its valid or not. but you
have given me a couple specifics I can certainly check
now. All pcb's get new crystals right away so they
usually are pulsing away.. One of those boards has a
new crystal pulsing but the cpu (verified) has no life
to it.. cannot remember which one but its probably a
good candidate for the clock section rebuilt. I just
wish I knew where to go chip by chip using the
schematics instead of the shotgunning..

If you are checking across a pcb, how do you know what
should be high/low/pulsing? I have a printout for all
74ls's that show inputs and outputs, but not every
single one is in use at every point in time I would
think.. I have schematics, and they show what pin goes
to what other thing, but I don't have a clue what it
'should' be at any point in time. Do you just start at
the cpu and find a pulsing line, then work backwards
through the chips (inputs to outputs) across the pcb
until it stops pulsing on an output basically?

Thanks!!

--- Jess Askey <jess@askey.org> wrote:

> andre wrote:
> > I have 3 Space Duel boards, with 3 issues..
> >
> > pcb 1) Has no life.. the -28 volt pin only reads
> about
> > half of that.. The 555 is new and both diodes
> beside
> > it test fine. There isn't much else in this
> circuit..
> > the electrolytics are new as well.. Any idea on
> what
> > else could cause the low reading? Can I replace
> the
> > 100 volt, type 100 or whatever they call it diode
> with
> > a 4xxx series?
> >
> I don't believe that the -28V effects anything
> except the EAROM so this
> shouldn't keep the board from dying. All you should
> need is a solid +5V
> and the 10.3V to get the logic up and running. You
> need the ±22V and all
> related voltages out of the regulators for the DAC's
> and sound. Make
> sure you have those and then you should checkout the
> reset pin on the
> CPU. Make sure it is low then stays high on powerup.
> If it is pulsing,
> then you have a watchdog barking, tie WDDIS to
> ground and see if you can
> get a display to give you error codes in self test.
> If the 10.3 is there
> and reset is behaving, make sure you have a solid
> clock (is it 12MHZ on
> SD?)... check on the main 12M test point to make
> sure it is there. If
> not, your crystal is most likely yanked or cracked
> (very common).
> > pcb 2) Has no life. The first eprom pulses fine,
> but
> > none of the other ones have any pulses on the
> pins. I
> > replaced the 244 immediately after the processor
> and
> > that did nothing. Even if the eproms were bad, it
> > would still pulse correct? (I am sure they are ok
> > though) What would cause only all the eproms after
> the
> > first one to be dead? The connections are fine as
> I
> > traced the pins for continuity..
> >
> >
> Check the stuff for above first and get a baseline.
> Then, I generally
> start by probing each address pin, you cant
> guarantee that they will all
> be pulsing because the program might be in a tight
> loop where it is only
> cycling the bottom two address bits. The lower ones
> should all be
> pulsing tho. Take out all the EPROM's on the CPU
> side except the top
> Address one (that is where Atari generally puts the
> Self Test routine),
> you can tell which one that is by looking at the
> owners manual and of
> the ROM's will have an * note saying that this ROM
> must be functional
> for self test to run. What you are doing is
> eliminating possible bad
> IC's that aren't 100% required for Self Test. Now,
> look at the data
> lines on the CPU, they should *all* be pulsing, if
> not you might have a
> shorted buffer or a shorted RAM holding one high or
> low. While you are
> at it, swap the CPU with a known good one to be sure
> yours is good (dont
> bend pins or you start to get into chip swapping
> hell). :-)
>
> > pcb 3) This board seems to work fine.. but the
> objects
> > flying around, or even your ships, sometimes
> shifts up
> > and down very slightly..but nothing else. I know
> on a
> > Star Wars a bad prom can cause only 'certain'
> vectors
> > to flake out while not affecting the rest, but on
> a
> > Space Duel what would cause that?
> >
> A failing Y counter on the lower bits might cause
> this. Not quite sure
> the best way to troubleshoot tho.
> > And one Gravitar...
> >
> > This one is pretty odd.. the pcb works fine.. Only
> > issue is if you shoot one of the red flying craft
> on a
> > planet, or tractor fuel, the sound for the reactor
> > (that you shoot in the reactor planet) will play
> > forever until you die or leave planet. The tractor
> > beam sound is there..but not the beeping of the
> fuel
> > you are picking up.. If you explode, there is no
> > explosion sound..but everything else explodes
> fine.
> > Self test shows nothing wrong and both pokeys were
> > swapped out.. and nothing.. any ideas?
> >
> >
> Hmmm... that *is* an odd one. Look for cut or
> shorted traces due to
> excessive PCB piling and abuse... you might be
> getting a really odd side
> effect of a pokey addressing issue maybe? The solder
> on the Atari PCB's
> is prone to being scraped and creating a trace
> short. I have seen really
> odd effects by random shorts and cuts on the PCB's.
> Otherwise, make sure
> the pokey address decoders are looking okay.
>
>
>
> hope that helps or at least puts you on a new set of
> wild goose chases. :-)
>
> jess
>
>
>
>
> > Thanks in advance.
> >
> >
> >
> >
>
____________________________________________________________________________________
>
> >
> > Be a better friend, newshound, and know-it-all
> with Yahoo! Mobile.
> > Try it now.
>
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
> >
>
---------------------------------------------------------------------------
>
> >
> > ** Unsubscribe, subscribe, or view the archives at
>
> > http://www.vectorlist.org
> > ** Please direct other questions, comments, or
> problems to
> > chris@westnet.com
> >
>
>

      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Tue May 6 09:23:17 2008

This archive was generated by hypermail 2.1.8 : Tue May 06 2008 - 21:50:00 EDT