Contents of the SEGA12.TXT file
Connecting the Sega 3D Glasses to the Macintosh
Part I, The Hardware
Version 1.2, Copyright 1989 Juri Munkki
This document describes how I connected the Sega 3D Liquid Crystal
Shutter Glasses to my Macintosh II RS-422 port. With the glasses you can
use software specifically made to display a stereo pair of images. This
method of displaying 3D graphics gives most users a feel of depth in the
Stereo viewing has interested me for years. I regularly read the usenet
rec.games.video newsgroup, since it occasionally contains information on
coin-ops. When the Sega Master system was introduced, I read about it
and discovered that it can control 3D-glasses. I immediately became
interested, but I had trouble finding Sega in Finland. This lead me to
post an article in rec.games.video. As it happens, Stan Lackey had
already built an interface for his 8-bit Atari computer. With his help,
I built an interface that works with the Macintosh. It should work fine
with any RS-422 or RS-232 serial port and if you know some electronics,
it shouldn't be too hard to modify it to work with other interfaces as
well. All my thanks go to Mr. Lackey.
Building the circuit requires a certain amount of experience with a
soldering iron and I recommend that you verify your circuit with a
digital multimeter. You are the only one responsible, if this circuit
manages to fry your serial port, your computer or something else. The
circuit can be tested without plugging it to the serial port. I
recommend that you use a 9V battery and connect it to the inputs that
would normally take the TxD+ and TxD- signals.
I haven't drawn a circuit board for this project. I simply used a
prototyping board and connected the components with wire.
Here's the list of components I used:
1CMOS-4013D-flip/flop with two inputs
1CMOS-4070Quadruple 2-input exclusive or gate
1RectifierI just bought one with a low voltage specification
20.1 uF capacitors
10.01 uF capacitor
23 kOhm resistorsI used a single 4.7 kOhm trimmer
13.5 mm stereo connectorA female connector like those in Walkmans.
1Circle-8 connectorA male connector for the Macintosh serial port.
someshielded cable with 3 wires
1VERO-boardA prototyping board of some sort
It is very important that you use CMOS-parts, since the operating
voltage is 10 V. We need 10 V, because the glasses use a 400 Hz square
wave when they are opaque. When they get 0 V, they are transparent.
Below is the circuit that I used. It's probably not optimal (I think it
might work even without the 4013, if you changed a few resistor values),
but I have built two of them and both work. The one I'm currently using
has a trimmer instead of the two resistors. If you are using a trimmer,
you might want to use an oscilloscope to verify that you are getting 400
Hz at the output plug or 1600 Hz at the 555 output (Pin 2). I used my
Mac II as the oscilloscope. Since the frequency is audible, I was able
to digitize it with my Impulse audio digitizer.
Figures are in a separate GIF file!
The rectifier (the leftmost component) is used to provide power for the
circuit. I connected the 0.1 uF capacitor to the rectifier output so
that the power would be steadier, but I haven't actually calculated what
it's effect is. I don't think that it has caused any problems, but
someone more proficient in electronics might want to put something else
The 555-timer is connected according to a circuit in a data book I found
here at the university. The frequency of this circuit is 1.44 / ((Ra +
2Rb) C), where C is the capacitance of the capacitor between pins 1 and
2. We see that with the specified resistors and capacitor, the frequency
is 1600 Hz. The D-flip/flop is used to divide this frequency to 400 Hz.
The 400 Hz clock signal is fed to the other input of three exclusive or
gates. Of the remaining three pins, one is grounded. That gate outputs
the clock unchanged and it is connected to the ground wire of the 3D
glasses. The remaining two pins are connected to TxD+ and DTR. These two
lines can then be used to invert the clock signal. When the clock is not
inverted, the output is the same as in the ground wire, i.e. the
potential is 0 V. When the clock is inverted, the we get 400 Hz
alternating current of +/- 10 V. The signals look like this:
Figures are in a sepate GIF file!
Here are the pinouts of the three circuits I used:
Figures are in a sepate GIF file!
The following three Macintosh serial port pins are used:
Pin 6 is TxD+
Pin 3 is TxD-
Pin 1 is DTR
The easiest way to solder the pins is to buy a connector where the pins
can be removed so that you can solder each pin individually. I also used
shrinking tube (I don't know what it's called in English) to make sure
that the wires do not touch each other.
This document will be followed with Part II, a brief description of the
software interface that is needed to program the glasses from a
Macintosh. I already have some animation and fast 3D transformation
routines ready, so I'll probably try to find time to write games for the
glasses. I'll make the games either shareware or public domain.
I'd really like to see a lot of software that supports these glasses and
I wouldn't mind if someone started selling these interfaces to "the rest
of us". If you want the sell the interface, please contact me before
selling more than a few. I think any profits should be shared with
those, who pioneered this interface. If you write software for the
glasses, a nice way to remember me is to send a free copy for me to try
out. I can then keep a list of compatible applications and make that
information available to users (and possible clients).
I occasionally come up with ideas like this interface and surprisingly
often have the energy to do something about them. If you want to
encourage me in these projects, please send me a postcard or letter.
Donations are also ok.
Helsinki University of Technology
Otakaari 1, U044B
This is version 1.2, which means that I'm no longer the only one who
has built the interface. Please destroy any older versions, if you
encounter them. There is a file describing the changes from 1.0 and 1.1
to 1.2. Version 0.9 had two capacitors in the wrong places.