Re: G08 deflection custom chip

On Mon, 29 Jul 2002 21:51:41 -0400, "Mark Jenison" <jenison@enteract.com> wrote:

>
>----- Original Message -----
>From: "Zonn" <zonn@zonn.com>
>To: <vectorlist@synthcom.com>
>Sent: Monday, July 29, 2002 6:57 PM
>Subject: Re: VECTOR: G08 deflection custom chip
>
>
>> On Mon, 29 Jul 2002 14:02:33 -0700, Zonn <zonn@zonn.com> wrote:
>>
>> Just to clarify, I'm not currently working on a Sega custom chip
>replacement,
>> but if anyone is interested in the above circuit, let me know.
>
>I'm am interested in your circuit. What would it take to get you to start
>working on a replacement Sega custom chip?? ;-)

Immortality!

Ok, maybe just a few 30 hour days would do it!

>Anyway, maybe some people could help me at least figure out the pinouts of
>this device?
>
>Pin 1 - wiper of height pot. Other inputs to pot are Y analog signal and Y
>yoke output
>Pin 2 - output that drives X circuit
>Pin 3 - wiper of height center. Other inputs to pot are -9.1V and +9.1V
>Pin 4 - GND
>Pin 5 - connected to width pot. Wiper is connected to X yoke output, and
>other input is NC
>Pin 6 - +9.1V
>Pin 7 - ? hooked to pin 10 via 12pf capacitor. According to manual, output
>should be 7.5VDC
>Pin 8 - NC
>Pin 9 - ? hooked to GND via 1K resistor
>Pin 10 - ? Y analog input
>Pin 11 - Y analog input
>Pin 12 - -9.1V
>Pin 13 - connected to pin 14 via 22k resistor
>Pin 14 - X analog input
>Pin 15 - connected to pin 14 via 22k resistor (same as pin 13)
>Pin 16 - ? Y input; hooked to pin 10 via 22pf capacitor and 30k resistor in
>parallel
>Pin 17 - wiper for V center; Other inputs to pot are -9.1V and +9.1V
>Pin 18 - output that drives Y circuit
>
>I'm not exactly sure what 7, 9, 10, and 16 are. If someone could take a
>quick look at the schematics to help me flush these out a bit? Schematics
>can be found many places online, but they are in my Electrohome G08 FAQ at
>http://www.enteract.com/~jenison/mars for sure.

These look like pretty typical connections if the part is used as a
amplifier/pincushion correction. Assuming internally the part uses some type of
transconductance amplifier to do the analog multiplications (I'm not sure how
else this would be done), then a bunch of bias currents must be set properly
which probably explains the different resistors going into different pins.

Shoe horning in a new design for this custom chip, into the existing circuit,
might be a bit of a pain. It's probably more worth while to design a new
deflection board. But we've seen how far that usually makes it. ;-(

Bill doesn't think our current ZVG schematic is ready for prime time (he's
probably right), but here's a link to the pincushion correction circuit we're
using:

http://www.zektor.com/misc/LM13700_pincushion.pdf

Both halves of the LM13700 are configured as a 4 quadrant multiplier. Currents
going into the '+-' inputs and the 'bias' input are multiplied and the output is
a voltage representation of the two signals multiplied. The LM13700 is similar
to a standard Op-Amp with two exceptions. 1) It gives you access to the
normally constant current source used internally to reference the two input
transistors (this is the 'bias' input), and they've added linearizing diodes to
allow more dynamic range with less distortion than would be available in a
normal transconductance amplifier (like the CA3020).

The output of the LM13700 amplifiers is very weak and two darlington buffers are
supplied to boost the current gain, but the 1.4v offset that results in using
these made them worthless for pincushion correction, also, the correction signal
needs to be inverted, so we added a couple of op amps.

Starting at the Y input, the Y signal is feed to both sides of the multiplier
creating a Y*Y signal. This signal must be inverted (U14A) and feed back into
the X correction multiplier.

The X correction multiplier pre-distorts the X-axis based on the position of the
Y axis using the equation Xout = Xin * (1 - Y * Y * K).

R41 sets the center of correction, so that a Y value of 0v does no correction.
R26 sets the amount of correction. This sets the 'K' in the above equation, it
also inverts the signal giving you the "-" part of the equation.
R19 adjusts an offset current that moves the correction out towards the edge of
the monitor. Without this a Y value of 0 would result in the an X value of 0
(anything times zero is 0). So this adjusts the unity gain of "1" part of the
equation. Theoretically these values are all unchangeable constants which is
why both Atari and Sega have no adjustments. For reliability, and ease of setup,
we're looking to replace these adj's with precision resistors in the ZVG.

Hope this helps anyone who might want to build a new deflection board, or have a
yoke designed.

Using this or a similar circuit might allow yokes to be made that don't need to
do pincushion correction, which might make them easier/cheaper to build, and/or
possibly work on more than a single CRT.

-Zonn
---------------------------------------------------------------------------
** 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 vectorlist-owner@synthcom.com.
Received on Tue Jul 30 15:07:08 2002