Steven,
Sneak current paths through the i/o lines are most likely the culprit. You
would be better off leaving all units powered up and just programming
one set at a time as outputs, and leave the other chip with the
lines as inputs.

You need some way to communicate the i/o status. If you have an i/o
line available on both devices (fat chance!) you tie these lines
together and pull it high with an external resistor or by programming
weak pullup. Initially each device uses this line as as INPUT. When
a device is going to take over the LCD lines it changes this line
to an output and asserts a LOW. It then takes over the LCD lines. At
completion of task it relinquishes LCD lines and changes this line
back to an input. Each device polls this line prior to attempting
a takeover of the LCD lines.

Personally, the approach that I would favor would be programming one
device to handle all communications with the LCD. One i/o line would
be a serial data input line that is 'shared' in bus fashion by as
many transmitting devices as desired. One output from the LCD PIC
would be used as a READY line to indicate to all transmitting devices
that it is OK to initiate a transmission. If you don't have the
extra output available, it IS possible for the transmitting devices
to monitor the serial data line. In this case a device begins by
checking to make sure that the serial data line has been high for more
that one frame time (time to send start, data, stop bits).
With either method you can program different devices to wait longer
amounts of time following the last LOW transition, which assigns
them a lower priority.

Hope this helps.
Fr. Tom McGahee


>
> At 14:56 1998-07-16 -0700, you wrote:
> >Friends,
> >
> >I am working on a project where I am using (2) 16C64A's and (1) 16F84. I
> have 1 of the 16C64A's going to the LCD's 10 data lines, the 16F84 shares
> these lines but are off (+5V Physically isolated) at different times. For
> some reason the 16F84 is interfering with the 16C64A when it (16C64A) is
> powered up. The 16C64A DOES NOT interfere with the 16F84 when it (16F84) is
> powered up. Any suggestions would be GREATLY appreciated! TIA
> >
> >                Best regards,
> >                Steven