A couple of thoughts. First, the LCD microcontroller should be on
the same PCB as the display. This allows changes to support a
variety of displays and interfaces.
The MIDI out signal is generated via two resistors so there's not a
lot of circuitry involved. Monitoring the TTL signal is appropriate.
There are two options here: simply monitor the MIDI out signal or
loop it through the microcontroller.
Advantages of looping it through the microcontroller is that you can
actually filter out the display sysex messages and not send them
out. Since they are sysex messages it shouldn't matter but you
always have the opportunity to add or expand the protocol since you
can filter the messages out. Right now I'm using the educational
ID. I suppose that is OK since this is a non-commercial
application. It doesn't really matter, though since the messages
are never passed through to MIDI out.
Disadvantages are that it adds a 1 or 2 byte delay to the message.
Pretty minimal.
One possibility would be to loop through the TTL signal at the
connector (e.g. put a jumper on the connector). This allows MIDI to
work without a display but when a display is connected, it filters
out the display messages. It can still be done with a 4 pin
connector.
6 of one, half dozen of the other ...
Dave
--- In
ComputerVoltageSources@yahoogroups.com, "Grant Richter"
<grichter@...> wrote:
>
> A complete MIDI interface will be on the main board.
> As I understand it, the LCD display microcontroller monitors the
> MIDI data stream looking for a SYSEX string.
>
> My thought was to simply send the 5 volt level TXD pin to the
> external LCD board. Not the current loop signal.
>
> That would eliminate any need for level shifting or an opto
> isolator on the display sub assembly.
>
> It would also allow the display to mounted in many more mechanical
> positions if seperate from the main board, and with it's own power
> supply.
>
> It would also be possible to have different size displays, or
> multiple displays each with a unique SYSEX code.
>