TI-H: IR Link


[Prev][Next][Index][Thread]

TI-H: IR Link



Hi,


I know Marc Levine has designed an IR Link for TI calcs but as far as I
know, the current design shouldn't work. Has someone built it and tested
it?


If it doesn't work, I might design a new one. I have no experience in
building IR devices and not much experience in electronics either, but I
think that with some help from this mailing list it would be possible
for me to build a working IR link for two or even more TI calcs (an IR
LAN in a classroom would be really great! :) ).


The first problem seems to be how to implement the IR link in the design
of the TI (I use a 85, but it should be possible to do other versions
later). It should probably be connected to the link port. The main
problem in IR is that you have only one channel (unless you want to
experiment with different wavelengths, but I'm not going to do that), so
you must somehow modulate the R and W signals in just one channel. I
don't know if there exist any useful protocols for this, but I have in
mind one way of doing it quite simply.


Then the hardware; I see two ways of connecting an IR LED and a receiver
(PIN diode?) into the link port. The "stupid" and easy way of doing it
would be to do a simple circuit with maybe one or a few transistors and
connecting it to the link port. This way the ROM routines for
transferring data between TI:s couldn't (AFAIK) be used, all the
programs would have to be run from ZShell. Also, you couldn't play games
(unless they start to support IR) over the link. The "smart" way would
be to connect some simple and cheap microcontroller in between the TI
and the IR LEDs and receivers. This would need some programming of the
chip(s) and cost a bit more, but wouldn't waste as much processor speed.
I have never used microcontrollers in my circuits, but I don't think
that's a problem. The problem is, is there enough reason to do a bigger
job and build the IR link with a microcontroller and spend more time and
money.


I hope I can find or think up a protocol for modulating the R and W
signals into the IR channel, hopefully so you can link more than two
TI:s with one link. Good thing the link protocol of the TI isn't
time-critical, so transmission errors should be easy to handle. On top
of the RW->IR protocol you could maybe even use the I2C or some similar
protocol to link many TI's, and maybe later other I2C devices like
EEPROM chips by using microcontrollers.


Then my questions:
Does this sound stupid?
Have I forgot something?
Any ideas?
Anyone willing to help?
Is there actually any interest in the idea or is everybody just
expecting a working radio link?


<pre>
-- 
*** Osma Suominen *** <a href="mailto:ozone@clinet.fi">ozone@clinet.fi</a> *** http://www.clinet.fi/~ozone
***
</pre>