vector monitors

From: Al Kossow <aek>
Date: Thu Aug 28 1997 - 11:21:27 EDT

Joel said:
"It sure would be nice to take an off the shelf audio amp
and use it as a replacement"

I've been talking to one of Apple's main analog desing
guys (who happens to sit in the next cube over from me)
about the design of the deflection circuits and he
showed me some of the problems with the deflection amps.

He just shook his head and rolled his eyes when I showed
him the G08 deflection amp design. The Wells designs are
pretty conventional designs (they just look like high-
power op amps) but the lack of overcurrent protection in
the output stage is a problem (which is patched by the
input protection circuits that were added)

The problem with these designs is the huge ammount of
power burned across the output transistors when the
deflection voltage is large. We guestimated >180 W of
power is being burned there. There is a current limiting
curcuit (and snubber diodes) on the later G08 boards
(the ones with the little paddle boards going to the
deflection transistors) but that doesn't do you any
good when the transistor shorts due to the junctions
melting

We measured the characteristics of the yokes yesterday
to try to see why the deflection rates would be better
on the G08, and to see what the DC and AC characteristics
were.

I've started to look at different designs for a class-D
MOSFET deflection amplifier design which is MUCH more
efficent (~90%) and would require almost two orders of
magnitude less power dissipated in the output transistors.

I'm also trying to see if the same design could be used
in the three XY color monitor designs. One of the problems
is trying to figure out what the little black box in the
G08 design is doing. If it's just an error amplifier for
the two channels (as the block diagram shows) or if there
is any X-Y correction going on in there as well, since
the Sega games have no VDR's or analog multiplier correction
circuits like the Atari games have.

One thing I'd be curious to see is the WG 6400 schematics
to see if they did anything different in the deflection
circuits.
Received on Thu Aug 28 08:21:33 1997

This archive was generated by hypermail 2.1.8 : Thu Jul 31 2003 - 23:00:55 EDT