🚀 go-pugleaf

In article <639@thunder.LakeheadU.Ca> dnwangus@flash.LakeheadU.Ca (Dave Angus) writes:
>In article <1992Dec14.223424.21680@unlv.edu> bill@pauli.lv-physics.nevada.edu (Bill ODonnell) writes:

>>     If extreme accuracy isn't a major concern you may
>>want to consider using the game port on the Apple.  The
>>Apple II series and the Apple IIGS both use a 4 channel
>>A/D converter to read the voltage (via a variable resistor)
>>at the game port.

>Sorry; that ain't correct.  The Apple ][ game ports work almost exactly
>like the IBM game ports:  they use a 558 quad timer chip to generate 
>time delays proportional to RC time constants.  The C is built into
>the game port, the R is provided by the joystick or paddles' 
>potentiometers.  The game ports can measure resistance, but not 
>voltage.

	There is a workaround for this problem; the timer requires
an input current, and a resistor and two PNP transistors makes
a pretty good voltage-to-current converter.

           +5V o------+----------+
                    |/         |/
Vinput+-----VVVVV-+-+----------+
             R    | |\         |\
                  +---+          |
                                 +(to game port)

The transistors should be the same type, and at the same temperature
(glue them together).  There is some slight temperature-dependence
due to the Vbe of the transistors, but it is possible to calibrate
the system for near-room-temperature conditions.  It's hard
to draw in ASCII, but the emitters of the two transistors
are connected to +5V (and they must be PNP).

	You want R to be roughly 220 k ohms for 0 to +5V inputs,
and 400 k ohms for -5V to +5V inputs.  The +5V power
is available, of course, at the joystick connector on the
game port.

	John Whitmore