Re: TI-H: Re: Ir Link compatability


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

Re: TI-H: Re: Ir Link compatability



Gregory Whitmore wrote:
> Then why does it work? Think about what you wrote, do it twice, the second time
> in reverse, what do you get? It works with the $serial link. i have not even
> used a graphilink cable.

OKay, I'm very sorry if I blame you for no reason. I'm not an electronics
specialist, but your IR link just seems too simple to work _with the TI
protocol_. If the link port worked so that one pin is used to send and the
other to receive, the IR link would work fine, but as far as I know it's a
bit more complicated than that.

I see quite a lot of problems with your IR link:

1. The TI protocol doesn't use just one wire for sending and the other for
receiving, but in fact both for them for both purposes, although not at the
same time. Your IR link has just one transmitter which is connected to just
one of the wires, so it can't know the state of the other wire. The IR
transmitter and receive modules send just one bit of data at a time, and
because the TI calculators have no accurate clock, one bit-wide data
channel without a clock or similar mechanism is not enough to send any
data. I didn't find a good explanation of the TI protocol to cite, but one
can be read in the Pascal code in the file
http://www.ticalc.org/pub/text/calcinfo/linkprot.txt

2. The $4 serial link, as pictured on all the sites where I have looked for
it (for example the two GIF schematics at
http://www.ticalc.org/pub/text/hardware), doesn't even connect to pin 3
(RxD) on a 25-pin serial port. Pin 2 (TxD) is not used for data but as a
reference voltage level. To make sure that we talk about the same link, the
$4 serial link I'm talking about is made by Timo Stenberg and Tobin Fricke.

3. I found the datasheets for the Sharp IR modules at
http://www.sharpmeg.com . The datasheets included an example circuit, which
was almost exactly the same as your IR link. But there was a crucial
difference, which I believe is a mistake you've made: The 7404 hex inverter
is on the TTL level side of the MAX223, not on the RS232 level side as in
your IR link. Your circuit might still work with the inverter on the wrong
side, however. I'm not sure about this, but it looks like a human error.


As I don't believe in magic in general, it is hard for me to accept that
your link works, even if you claim it does. But this discussion puts us
both in an irritating situation. I wouldn't like to be the one of us who's
wrong and I don't enjoy blaming people, but I like to know how a
(relatively simple) device works, and when it seems to contradict most of my
electronics and TI knowledge, I become a skeptic.

Could you tell me why you chose just the parts that you've chosen,
particularly why you used both a MAX233 (which is a TTL-RS232 voltage level 
converter) and the $4 serial link (which does the same) and connected them 
in series? As far as I know, you could have hooked the IR modules directly 
to the link port pins as well (possibly with an inverter in between).

-Ozone

--
*** Osma Suominen *** ozone@clinet.fi *** http://www.clinet.fi/~ozone ***


Follow-Ups: References: