atn@cory.Berkeley.EDU (Alan Nishioka) (09/21/89)
For a friend's mechanical engineering project I needed a circuit that latched. I built a circuit using two SPDT relays in which power through one coil closed a switch which energized the second coil which closed a switch which energized the first coil, or something like that :-) Kinda like an RS latch. This worked fine for my application, except there was a condition where if the wrong switch was closed, the system would oscillate. Does a simple device exist which does this sort of thing? (ie. Is there a better way to build a circuit with memory? (perhaps with transistors)) I didn't want to use a 7474 for simplicity and power supply reasons. At the time, I looked at some of the "latching relays" I had heard about. These apparently have two coils, and a pulse to one coil latches the switch one way, while a pulse to the other coil latches the switch in the other state. This wouldn't have been useful to my application. Alan Nishioka atn@cory.berkeley.edu ...!ucbvax!atn%cory
erk@americ.UUCP (Erick Parsons) (09/22/89)
From: atn@cory.Berkeley.EDU (Alan Nishioka) Message-ID: <17419@pasteur.Berkeley.EDU> >For a friend's mechanical engineering project I needed a circuit that >latched. I built a circuit using two SPDT relays in which power through >one coil closed a switch which energized the second coil which closed a >switch which energized the first coil, or something like that :-) Kinda >like an RS latch. This worked fine for my application, except there was >a condition where if the wrong switch was closed, the system would >oscillate. A very common latching circuit (for motor control) is to use one relay with two sets of contacts. It is a DPDT relay. ^^^ It requires three wires in the control conduit, two switches: SPST momentary contact, one normally open and one normally closed. 1st set of contacts for control _________ Hot | | ___ | ------o-----------|/|-----------o-------------- . . --o | | | | o-----o-----------| |-----------o-------------- .---. --o | |_________| | | C |-----o------------O------------o-----------------------o I | L Gnd _________ | | o-----------|/|-----------o | | o-----------| |-----------o |_________| 2nd set of contacts for load -----------symbols used------------ |/| normally closed contact of relay | | normally open contact of relay .---. normally closed momentarily opened manually operated switch ___ . . normally open momentarily closed manually operated switch -- ------------------- // ---------------------------------------------------- Erick Parsons // Knowledge is little more than knowing the questions ------------- // erk@americ.UUCP or ..ames!pacbell!sactoh0!americ!erk Sacramento Ca \\ // GEnie: E.PARSONS %-) Hamatuer Radio N6RZB -------------- \X/ --------------------------------------------------------
larry@kitty.UUCP (Larry Lippman) (09/24/89)
In article <2030.AA2030@americ>, erk@americ.UUCP (Erick Parsons) writes: > A very common latching circuit (for motor control) is to use one > relay with two sets of contacts. It is a DPDT relay. ^^^ > It requires three wires in the control conduit, two switches: SPST > momentary contact, one normally open and one normally closed. > _________ Hot > | | ___ | > ------o-----------|/|-----------o-------------- . . --o > | | | | > o-----o-----------| |-----------o-------------- .---. --o > | |_________| > | > | C > |-----o------------O------------o-----------------------o > I | > L Gnd > 2nd set of contacts > for load The above scheme is inadvisable, and simply will not work with many relays. The reason is that most relays intended for power control purposes have a SPDT contact configuration known as "break-make" (BM). Operating the NO "on" switch will energize the coil, but as the armature begins to pull in, the break contact will break, interrupting the current flow to the coil BEFORE the make contact will make to latch the relay on. The net result is that the relay will chatter, but never pull in to a fully operated and latched state. While a contact configuration of "make-break" (MB) will permit the above circuit to be functional, such a MB contact arrangement is relatively uncommon in a power control relay. In reality, running the "on" switch with a break relay contact serves no useful purpose. The above circuit may be used if the "on" switch is connected directly to the relay coil, with the break contact being unused. A motor contactor, for example, uses a single make contact for the latch circuit. While the heyday of relay logic control circuits has now passed due to the advent of solid-state logic, microprocessors, and industrial Programmable Logic Controllers (PLC's), relay contact timing can still be a problem if not considered in the design of some control circuits. The concept of relay and relay contact timing was honed to a fine art in the days when telephone switching circuits were solely designed using electromechanical technology. As an example, a common style of of wire-spring telephone relay having TWELVE contact sets could be configured with any combination of make (M), break (B), make-break (MB), break-make (BM), early make-break (EMB), early break-make (EBM), early make (EM) and early breek (EB) contacts! In addition, not only could such a relay be designed with up to three separate windings, but it could be designed using copper slugs as an additional one-turn winding to effect either a "slow-release" or "slow-operate" relay with effective "slow" timing as great as 250 milliseconds. I'd like to make one other point as long as I am here on my soapbox. :-) I do not fault the above poster for his circuit suggestion, even though it is generally unworkable. At first glance, the circuit seems reasonable and it seems that it should work. Many people seeing such a circuit would scoff at the suggestion of actually building and testing it since it seems "so simple". The point is, the REASON why this circuit won't work is NOT something that is taught today in most BSEE curricula. It is something that an engineer learns only through first-person hands-on experience. And such experience is something which takes time, involves making mistakes, and is not learned in any classroom. <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?"