Re: TI-H: Modems, Chips, and TIs...OH MY!!!
[Prev][Next][Index][Thread]
Re: TI-H: Modems, Chips, and TIs...OH MY!!!
>Mel Tsai wrote:
>
>> But a schmitt trigger probably won't speed up transfers, merely just
>> make the output cleaner. It will still have slow rise and fall times,
>> I doubt it will go any faster even with a cleaned-up signal.
>
>Not so, I think. If you use a Schmitt trigger you don't have to wait for
>the voltage to drop to about 1V or rise to about 3V (TTL) before
>changing the state of the port. You can set the trigger level to about
>e.g. 2,5V, so 2,3V will result in a 0V output and 2,7V will become 5V.
>If you follow the TTL specs you must take the signal down to about 1V
>before allowing it to rise again (which takes time when there is a
>capacitance), but keeping the level at about 2,5V will allow you to send
>data faster.
Okay, I understand what you mean. But how do you get a schmitt
trigger to reference to these voltage levels? The problem is that
you'll have to create a precise reference so that it will trigger at
(for example) 1V and 3V.
Another (huge) problem is that there's absolutely no way to guarantee
that the link will go above and below those levels within a single IN
7,A and OUT 7, A instruction cycle, and the timing on this will be
very difficult, probably impossible. There's no easy way to cheat
capacitances! But if you can figure out exactly how to organize this
then it may speed transfers up dramatically (but it will complicate
the expander design due to the extra schmitt trigger chip).
-Mel
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-The TI-Memory Expansion Homepage
-http://www.egr.msu.edu/~tsaimelv/expander.htm
Follow-Ups:
References: