Relay computers - OR functionality

From: Dwight K. Elvey <dwight.elvey_at_amd.com>
Date: Mon Sep 27 18:21:10 2004

Hi
 I thought I'd show a case where one might want to
use diodes.

 In my logic, 'A' would indicated the normally opened contact
an 'a' would be the normally closed contact for relay A.
I'll use + to indicate OR and . to indicate AND.

Lets say we wanted to create the equation:

 X=(A.B)+(C.D)
 
 We can stack the normally open contact of A in series
with a B contact. We can then parallel these with the series
of C and D. The output of these can be tied to the
coil of X.
 Now say we also needed:
 
 Y=(A.B)+(E.F)
 
 We would have to used a new set of contacts for the A.B term
because the wires work both ways. If we were using DC relays,
we can add a pair of diodes in series with the two AND'd
terms used to drive each X and Y. This reuses the same contacts
to drive both X and Y.
 So, as you can see, the diodes are not necessary but can
be used to save on the number of isolated contacts needed.
 One can also play boolian games with these equations to
do things like reduce total power required. Since each relay
is assumed to have both inverting and non-inverting outputs,
one can change where things are inverted:

 y=(a+b).(e+f)
 
 If the case is that Y would be more likely to be active,
this second equation makes more sense for power.
You can see here that we have optimized the power without
doing anything other than change a few wires. We may
have also minimized the need for extra contacts and or diodes.
 Like I said, designing with relays takes a different way
of looking at the problem. One can also see that the OR
function can be created by putting contacts in series or
in parallel, depending on the active state of the relay that
one is driving. The same goes for the AND function.
 This adds another flavor to things because one can move which
side of the relay one places the term to optimize, power,
delay or number of contacts used.
Dwight
Received on Mon Sep 27 2004 - 18:21:10 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:37:32 BST