2 Jan
2006
2 Jan
'06
9:49 a.m.
Anyone have intermediate to advanced level advice on relay logic?
When I was a kid, each relay would cost about one hour of labor, a cheap
bench linear supply could only handle about 10 active relays, and fanout
ratios were maybe only 5 to 1.
Now according to my latest Mouser catalog, each relay will cost me about
two minutes of labor, a cheap switching power supply will run hundreds of
relays, and I can easily get a 10 to 1 fanout even with cheapie relays.
I'm quite well aware of how to make basic gates, FF, counters etc out of
relays but I'm curious about anyone elses experiences. I can contemplate
all kinds of weird problems like mechanical shock from so many relays
clicking at once, or the relay equivalent of contact bounce, but I don't
know if they're real problems. Then there's mounting issues.
3 Jan
3 Jan
4:20 p.m.
I'm quite well aware of how to make basic gates,
FF, counters etc out of
relays but I'm curious about anyone elses experiences. I can contemplate
all kinds of weird problems like mechanical shock from so many relays
I've never heard of that being a problem. None of the books I have on
telephone exchanges (which used thousands of relays) mention it.
clicking at once, or the relay equivalent of contact
bounce, but I don't
Countact bounce shouldn't be a problem either (well, maybe if you're
interconnecting very different relays, trying to control a high-speed
reed relay from a much slower type thing...). The response time of the
driven relay won't be fast enough to respond to contact bounce.
Be warned if you look at some of the telephone exchange circuits that
they used (at least in the UK) some odd relays. Not only
make-before-break changeover contacts, but also slow-energise and
slow-releasy relays (copper 'slugs' in the coil was one way to get this
IIRC). You may have problems finding that.
Incidentally, I made a relay flup-flop without realising it a few months
back. I had a pair of contactors (high-power relays, basically) that were
mechnaically interlocked so they couldn't both close together and
electrically interlocked so that the 2 coils couldn't be energised at the
same time. The latter was done by feeding each coil through a
normally-closed contact on the other contactor.
Anyway, if I turned on the supply to one contactor (let's call it 'A'),
it pulled in. Turning on the supply to the other one ('B') did nothing
(since the current path to B's coil was broken by the NC contact on A).
But now if I turned off A's coild supply, A dropped out and B pulled in.
I could then turn on the supply to A again and nothing happened until I
momentarily opened the supply to B.
Of course this is exactly what was supposed to happen and I fairly
quickly relaised it was logically equivalent to a pair of cross-coupled
NOR gates -- the classic Eccles-Jordan circuit.
-tony
7:01 p.m.
Incidentally, I made a relay flup-flop without
realising it a few months
back. I had a pair of contactors (high-power relays, basically) that
were mechnaically interlocked so they couldn't both close together
and electrically [mutually inverting-coupled].
I fairly quickly relaised it was logically equivalent
to a pair of
cross-coupled NOR gates -- the classic Eccles-Jordan circuit.
Actually, cross-coupled NAND gates - unless you count "power present"
as a logic 0, rather than 1 - because the "maintain memory" state is
power present on both supply lines and the "lost memory" state is power
absent on both, rather than the other way around.
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
8:03 p.m.
Relay logic can depend on more than simple make/break on
coil-current-present types.
Stepping relays were used in telephone and logic systems. I used to have a
telephone relay with two windings on the same electromagnet. I'm not sure,
but I suspect the two might have been used as sort of a logical "AND"
function. Latching relays use multiple coils also.
Cheers,
Chuck
4 Jan
4 Jan
7:39 a.m.
From: "Chuck Guzis" <cclist at sydex.com>
Relay logic can depend on more than simple make/break
on
coil-current-present types.
Stepping relays were used in telephone and logic systems. I used to have
a
telephone relay with two windings on the same
electromagnet. I'm not
sure,
but I suspect the two might have been used as sort of
a logical "AND"
function. Latching relays use multiple coils also.
There are several reasons for multiple relay windings, especially on
telephone relays:
two windings, where the sum magnetic field of both is required to operate
the relay - often used in off hook detect circuits (a and b wires both go
through relay, inductance of coils prevents speech from entering the
exchange power supply).
two windings, where one will operate the relay in a normal manner, the other
provides enough flux to hold the relay in - used in some decoders.
three windings, two operate the contacts, the third is used to couple tones
(NU, dial, engaged) to the speech pair.
there are lots of other configurations - these are the most common!
slow operate relays have a copper "slug" on the core (basically a large
shorted turn which slows the build up of flux).
slow release can either use a slug, or a resistor in parallel with the coil
(slows the collapse of the magnetic field). a longer delay can be obtained
using a diode in a similar way.
If you want the theory of relays, try Atkinson's "Telephony" (long out of
print, but many libraries have copies still).
Jim.
4:57 p.m.
If you want the theory of relays, try Atkinson's
"Telephony" (long out of
print, but many libraries have copies still).
This is an excellent set of books (normally stated to be 2 volumes, there
is a suplement to cover some 'new' exchanges, which is even harder to
find), they were the books that finally got me to understand the
telephone system when simpler books had failed (mainly because the latter
had missed out all the good stuff).
Volume 1 covers the subscriber sets (the telephones themselves) and
manual swtichboards, Volume 2 covers automatic exchanges. The latter is
the one you really want, although you probably should read them both.
They do turn up second-hand from time to time (I've even found them in
charity shops!), but they are not that common.
-tony
4:37 p.m.
Relay logic can depend on more than simple make/break on
coil-current-present types.
Stepping relays were used in telephone and logic systems. I used to have a
telephone relay with two windings on the same electromagnet. I'm not sure,
but I suspect the two might have been used as sort of a logical "AND"
function. Latching relays use multiple coils also.
Mutlipe windings on telehpone relays can have many functions, including
acting as an inverter (pass a constant current through one winding, the
relay pulls in, pass the same current through the other winding in the
opposite direction, the magnetic fields cancel, it drops out), provinding
2 inputs to the relay (either or both coils energised in the same
direction will make it pull in), and even acting as an audio transformer.
-tony
4:35 p.m.
Incidentally, I made a relay flup-flop without
realising it a few months
back. I had a pair of contactors (high-power relays, basically) that
were mechnaically interlocked so they couldn't both close together
and electrically [mutually inverting-coupled].
I fairly quickly relaised it was logically
equivalent to a pair of
cross-coupled NOR gates -- the classic Eccles-Jordan circuit.
Actually, cross-coupled NAND gates - unless you count "power present"
as a logic 0, rather than 1 - because the "maintain memory" state is
6879
days inactive
6882
days old
7 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
jim＠g1jbg.co.uk -
mouse＠Rodents.Montreal.QC.CA -
vlm＠mulhollon.com