Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 beta 3/9/83; site sdcrdcf.UUCP
Path: utzoo!linus!philabs!sdcsvax!sdcrdcf!darrelj
From: darrelj@sdcrdcf.UUCP
Newsgroups: net.micro
Subject: Re: Apple //e braindamaged keyboard
Message-ID: <350@sdcrdcf.UUCP>
Date: Sun, 26-Jun-83 13:57:43 EDT
Article-I.D.: sdcrdcf.350
Posted: Sun Jun 26 13:57:43 1983
Date-Received: Thu, 30-Jun-83 11:30:47 EDT
References: <108@iwu1c.UUCP> <915@uwvax.ARPA> <203@packet.UUCP>
Reply-To: darrelj@sdcrdcf.UUCP (Darrel VanBuer)
Organization: System Development Corporation--a Burroughs Company
Lines: 14

Part of a full N-key rollover keyboard implementation requires that the
connection made in the scanning matrix (typical scanning chips have the
potential for somewhere between 70 and 110 keys in a not too far from
square array, e.g. 8x11) have a nonzero connection impedence under some
characteristics.  The possibilities include plain (0 ohm) switches with
diodes (so switch is unidirectional) and switches which create a resistor
or capacitor with measurable impedance, thus sneak paths have at least 3X
the impedance, so a carefully tuned threshold detector won't see false keys.

Another approach in some Hall-effect keyboards (in which each key contains
a little detector chip) is that the key sends simultaneous brief pulses onto
it's two lines in the scanning matrix on key closure.  Ghosts then can occur
only in the event two keys are pressed in one pulse time (less than a
millisecond, usually).