12-volt matrix input->output?
in [Design]
|
From: Tony Williams on 27 Feb 2007 03:24 In article <eru3fq$hb5$1(a)jasen.is-a-geek.org>, jasen <jasen(a)free.net.nz> wrote: > >> Switches Lamps > >> A B C W X Y Z > >> ----- ------- > >> 0 1 0 1 0 0 0 > >> 0 1 1 1 1 0 0 > >> 1 0 0 0 0 1 0 > >> 1 0 1 1 0 1 0 > >> 1 1 0 1 0 0 1 > >> 1 1 1 1 1 1 1 > > > > If you are already using relays then I think this can > > be done with 3-off 4-pole changeover relays. > overkill... smaller relays are sufficient I guessed it could be simplified but was too lazy to try it. Nice try, but could you check your ascii diagram for typos, there seems to be something wrong. eg, Look at the req'd '010' input, '1000' output. The B-relay is closed, connecting B1-NO and B2-NO. This does not produce a 1 on the W output? There are other anomolies. Contact C3 shorts Y and Z together, but look at the '101' input, this requires Y and Z to be different. There's a typo or I'm not reading your circuit correctly. > +---------- W output > | > .-------------------|------. > NC | NC | NC > / | / | / > / | / | / > 12v---A1 | 12v---B1 '---C1 > | > NO---| NO----+----------NO > | | > | | > | | NC > | | / > | | / > | `------C2 > | > | NO---> X output > | > | > | NC--------------NC > | / / > | / / > +-------B2 .-C3 > | > NO---+----|----NO > | | > | `--------> Y output > | > | > | > | > | > `---------------> Z output > /|\ /|\ /|\ > | | | > A-contacts B-contacts C-contacts > A-----. B-----. C-----. > | | | > ___|___ ___|___ ___|___ > | | | | | | > | CoilA | | CoilB | | CoilC | > |_______| |_______| |_______| > | | | > 0v----+---------------+---------------+ -- Tony Williams.
From: Rich Grise on 27 Feb 2007 21:03 On Sat, 24 Feb 2007 12:39:20 -0800, Joerg wrote: > John E. wrote: > >> I want to turn on 4 lamps in 6 combinations. These are controlled by 3 >> switches in 6 corresponding combinations of contact closings. >> >> The truth table looks like this >> >> Switches Lamps >> A B C W X Y Z >> ----- ------- >> 0 1 0 1 0 0 0 >> 0 1 1 1 1 0 0 >> 1 0 0 0 0 1 0 >> 1 0 1 1 0 1 0 >> 1 1 0 1 0 0 1 >> 1 1 1 1 1 1 1 >> > > http://focus.ti.com/lit/ds/symlink/cd4028b.pdf > > Digikey has it in stock even in the old DIP version, 52 cents a pop. If > you need more than a milliamp or so you could use a driver from the ULN > series. 74AC138, and some diodes: 138 outputs: A B C D 000 | | | | --------------------------- 001 | | | | --------------------------- 010 | | | | --------------X----X----X 011 | | | | -------------------X----X 100 | | | | ---------X----X---------X 101 | | | | --------------X---------X 110 | | | | --------------X----X------ 111 | | | | -------------------------- Where "X" is one diode of a diode OR, connected the usual way; the unused inputs default to 1111 for the output, since the 138 is active low. To make the default 0000, just strap all of the unused outputs with all four diodes. Have Fun! Rich
From: Rich Grise on 27 Feb 2007 21:04 On Sun, 25 Feb 2007 13:44:40 -0800, Joerg wrote: > John E. wrote: >> Thus spake Joerg: >> >>>As others have suggested this can be done using a decoder and diode to >>>wire up the desired lamp combination. Instead of analog muxes you can >>>also use the CD4028 and use only the A, B and C inputs. Wire D to ground. >>> >>>http://focus.ti.com/lit/ds/symlink/cd4028b.pdf >> >> I've looked at the data sheet. Fewer pins to deal with (no inhibit, no >> input/output control) -- looks more straightforward... > > Yep, keep it simple :-) > > BTW we just brought a couple back from church and his name is the same > as yours. He is a retired mineral engineer. I told him and he said that > he never met another person with that name. At least not in the last > nine decades or so and this guy has been all over the world. Not even Johnathan E from "Rollerball"? ;-) Cheers! Rich
From: Rich Grise on 27 Feb 2007 21:32 On Wed, 28 Feb 2007 02:03:28 +0000, Rich Grise wrote: > On Sat, 24 Feb 2007 12:39:20 -0800, Joerg wrote: >> John E. wrote: >> >>> I want to turn on 4 lamps in 6 combinations. These are controlled by 3 >>> switches in 6 corresponding combinations of contact closings. >>> >>> The truth table looks like this >>> >>> Switches Lamps >>> A B C W X Y Z >>> ----- ------- >>> 0 1 0 1 0 0 0 >>> 0 1 1 1 1 0 0 >>> 1 0 0 0 0 1 0 >>> 1 0 1 1 0 1 0 >>> 1 1 0 1 0 0 1 >>> 1 1 1 1 1 1 1 >>> >> >> http://focus.ti.com/lit/ds/symlink/cd4028b.pdf >> >> Digikey has it in stock even in the old DIP version, 52 cents a pop. If >> you need more than a milliamp or so you could use a driver from the ULN >> series. > > 74AC138, and some diodes: > > 138 outputs: A B C D > 000 | | | | > --------------------------- > 001 | | | | > --------------------------- > 010 | | | | > --------------X----X----X > 011 | | | | > -------------------X----X > 100 | | | | > ---------X----X---------X > 101 | | | | > --------------X---------X > 110 | | | | > --------------X----X------ > 111 | | | | > -------------------------- > > Where "X" is one diode of a diode OR, connected the usual way; > the unused inputs default to 1111 for the output, since the 138 > is active low. To make the default 0000, just strap all of the > unused outputs with all four diodes. > As long as I have it right here, http://focus.ti.com/lit/ds/symlink/cd74ac138.pdf This could be built on a protoboard with vertical diodes and flying wires across their tops. :-) Ooops! It's a 5V part. I guess the 4028 will do it, but the outputs aren't all nicely lined up like on the 138, so it'd be harder to do on a protoboard. :-) Cheers! Rich
From: Joerg on 27 Feb 2007 22:09
Rich Grise wrote: > On Sun, 25 Feb 2007 13:44:40 -0800, Joerg wrote: > >>John E. wrote: >> >>>Thus spake Joerg: >>> >>> >>>>As others have suggested this can be done using a decoder and diode to >>>>wire up the desired lamp combination. Instead of analog muxes you can >>>>also use the CD4028 and use only the A, B and C inputs. Wire D to ground. >>>> >>>>http://focus.ti.com/lit/ds/symlink/cd4028b.pdf >>> >>>I've looked at the data sheet. Fewer pins to deal with (no inhibit, no >>>input/output control) -- looks more straightforward... >> >>Yep, keep it simple :-) >> >>BTW we just brought a couple back from church and his name is the same >>as yours. He is a retired mineral engineer. I told him and he said that >>he never met another person with that name. At least not in the last >>nine decades or so and this guy has been all over the world. > > > Not even Johnathan E from "Rollerball"? ;-) > To be honest I don't know who that is. -- Regards, Joerg http://www.analogconsultants.com |