Re: WG 6100 Final Solution?

On Mon, 15 Oct 2001 17:07:38 -0500, tom mcclintock <tomm@mgcap.com> wrote:

>Guys,
>
>Thanks to Matt McCullar, I was able to get a spiffy version of this
>article from Star*Tech Journal. I'd been spending some time reading back
>through Gregg's doc, and the schematics, and all the other data on
>Gamearchive, and was wondering if anyone had actually performed some/all
>of the MODs outlined in this document?
>
>The most interesting bits (to me) were the low voltage section MODs...
>Anyway, my question here is - has anyone actually made any of these
>modifications/improvements and if so, how well do they work? I can't say
>that I've heard from anyone on the vectorlist that has done them, and
>they certainly are not listed in Gregg's article.

When I got my first Tempest I tried many of these things (before knowing about
this article.)

Things I discovered:

* Though it claims otherwise, removing the regulator from the low voltage side
make for irritating jitters and blooming on the vectors. This might not be so
bad from a "operators" point of view, but as a collector is was unacceptable.

* Removing the regulator increases the voltage across the output yoke drivers.
This might have an advantage of a slight increase of vector speeds when moving
large distances, but it also means the transistors must dissipate more heat
during large movements. Since Tempest was written for the monitor as it was
designed, the speed increase is meaningless, but the heat dissipation is
critical!

* Moving the transistors seems more like superstition to me. Aluminum is a
better conductor of heat than steel, so it could be argued that the X drivers
should remain where they are since the aluminum can dissipate heat faster than
the steel frame. If things need cooling, install a fan.

The low voltage regulator designed by Atari really sucked, I came up with a
re-design that was slightly more complicated than the LV2000 -- I switched to
the LV2000. The best thing to do with the low voltage section is to install an
LV2000.

* Limiting the output current on the output transistor may not be a bad idea, if
it does not slow down drawing speeds enough to affect the display. Though it
seems silly just to do the X transistors. During high speed moves, both X and Y
channels will have to pass the same amount of current. While it's true that the
X channel is larger than the Y, and therefor, on the average, will have longer
moves, and therefor possibly get hotter, the instantaneous current peaks on both
are the same. Since it's most likely these peaks that would blow the
transistors, you'd think both X and Y would be protected.

The experiments I did on the high voltage where different than those described.

One thing I noticed about the high voltage sections is that transistor Q900 is
dropping 19.6 volts! Based on a 2.1v drop across R901, Q900 is passing .538
amps, at 19.6 volts that's ~10 watts. (This could be a bit high since R900
supplies more than just Q900 with current.)

That was quite a bit, and since I didn't see anything else that needed the full
25v, except for T901, (It looks like they run the output HV transformer with
~30v), I installed a pre-regulator on the +V side. Since they were using a 13v
zener as a voltage reference, I chose a 15v regulator (7815 or equivalent).

This was a few years ago, but I remember it working pretty well. The heat
dissipation on Q900 went *way* down (everything ran nice and cool), and I don't
remember any adverse effects on the HV output (vectors were fine and nice and
stable.). With a +15v pre-regulator, Q901 would only be dropping 6.9v, or less
than 4 watts. This also significantly drops the current demands on Q901.

I don't currently do this since is was just "tacked on". I got really busy at
that point and never pursued it any further, so I never tested things like the
Star Wars explosion, etc. -- mostly just the tempest attract mode. So there
could be some unknown problems with this. It is also very possible that some
other resistor values might need changing in the voltage regulator to compensate
for the lower input voltage. (Lowering R905, R906 and R913 come to mind. And it
might be a good idea to re-route R917 to a position in front of the
pre-regulator.)

By dropping the voltage with a pre-regulator you move much of the heat
dissipation to the regulator and away from the HV regulator. This seems like a
good idea and should probably be investigated further.

Hey Anders, how about an HV2000? (I'd certainly buy a few!).

-Zonn

>Here's the doc with the 'new' schematics drawings thanks to Matt
>McCullar. You can actually read them!
>
>http://www.ionpool.net/final_solution.pdf
>
>
>Let me know if anything should be added or corrected. Thanks.
>
>tom
>---------------------------------------------------------------------------
>** 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 Oct 15 17:17:46 2001