Several months ago, we converted LETA to a CMOS Gate array and called it
LETA II. The old HMOS part is no longer available. It works fine (I think)
in Trackball/opto-coupler applications. Apparently, there are additional
circuit precautions needed if you plan to use it to read switches - the
fast transition of a switch together with the long control harness, causes
latch-up....see below...
...........................................................................
From:	SANDY::PAAUW        29-MAY-1987 13:41
To:	DOWNEND
Subj:	leta II

Ed said that the program would crash due to the Leta. I checked with Doug and
he said that the switch inputs were ringing below ground, probably due to the
length of the cables.  This can cause SCR latchup in CMOS and this was apparently 
the case with the Leta's.  The Leta chips would get very hot until power was
removed.  This is a classic symptom of SCR latchup.
  Doug replace the Leta with a lsttl part that also wouldn't work.  The inputs
would occaisionally be asserted onto the data bus even when the chip was not
enabled.  This would cause the program to crash. A Schottky part seems to be
able to handle the ringing and is working OK for now.  Doug believes that
the cure for the ringing would be to place series resistors in series with
the input wires.
  In any case, I believe that the problem goes farther than the Leta and that
SCR latchup is not surprising under these circumstances.
  I just realized that I'm sending this only to you. Could you forward this
to the others and to myself? Thanks.
-- Don