I have a 4-switch, 4-channel remote control control that activates four spdt relays. The device can be configured so that any of the four channel’s relays will “turn on” with one push of the button and “turn off” with the next push, etc. Alternatively, the four channels may be set as “momentary” such that its relay only stays active so long as you hold the button down. Releasing the button releases the relay. These units can be configured in one mode or the other, but not both simultaneously.

But I want to run three channels in the “momentary mode,” and one in “on-off mode” so needed to devise a simple circuit to accomplish this. IOW, I needed a circuit that could convert a “momentary on” to a “fixed on” with one push, then “fixed off” with the next push, alternating accordingly.

Sounds easy enough, and there are a ton of circuits on the net to convert a “momentary on” to a “fixed on.” But in these circuits, the “off” is accomplished using a separate switch (or second relay) to cut the power to the relay…, and that just wasn’t going to cut it for me. (I actually do a similar circuit on my railroad for my station “leap frog” but there are separate reed switches for the on and off in the two sets of tracks.)

This is how it is typically done. Note the separate “on” and “off” switches. But I need it to turn on and off using the same switch. Hmmmm…

More to come.

OK, back to it…

First the “ground rules.” This had to work with no more than two, 12 volt, dpdt relays, and any capacitors and resistors I had at hand. Therefore, all relay “steering” had to be done using minimal resources.

I came up with the schematics below.

and,

The remote control relay is just that…, part of the remote control system.

The remote control relay provides continous current as long as its button is held. This needs to be changed to a pulse of limited duration. If the current continues to be applied, the system will go into oscillation ending up in whatever state it just happened to be in when the button is released. Not good.

There are two easy ways to make a timed pulse from an intermittent source using a capacitor. One method is to run the current through the capacitor. It will conduct as it charges with the timing based on the mfd of the cap. A bigger cap will pass a longer pulse. But once the cap is charged, it needs to be discharged before it can create another pulse. I do this in the first schematic.

The second method is to charge the cap, then let the cap discharge, creating a pulse. Again, the timing of pulse is dependant on the mfd. This is shown in the second schematic. This also seemed easier to work with.

The idea is to first steer this pulse to one relay, then use that relay to steer it to the other. Current flows from the remote control relay to the center relay where it charges the capacitor as it lifts the armature. This capacitor then gives the armature the current necessary to make the trip to the other contact and depending on its size, will hold it there for a moment. Without the cap, this relay would just go into oscillation as it makes/breaks.

Once the armature of the center relay reaches the other contact, it energizes the relay on the right. When that armature lifts off the seat, the cental relay looses its power and and turns off.

So, we only want enough current to flow for a period long enough to energize the center relay, and what current is left over when the armature reaches the other contact needs to be low enough such that the relay on the right doesn’t lift from its seat. The capacitor on this relay serves as a reservoir that can absorb the left over energy without lifting the armature from the seat. Because the armature doesn’t lift, the center relay stays powered and we have a “latched” source of current to play with.

When the next pulse comes though, it is directed to the second relay making the armature lift from the seat de-energizing the center relay. Any left over energy should not be enough to power the center relay, that is absorbed by its cap, and the latched power unlatches.

THAT’S THE THEORY.

It then becomes a juggling/balancing act sizing the caps to accomplish the goal. Recognize that any little “twitch” of the right relay will deactivate the center relay, and I never was able to get this to work in a consistent manner, even after countless hours of parts substitution.

I was about ready to give up on it when someone on the other forum thanked me for a circuit that I came up with that de-powers the Sierra sound systems when used with supercaps before the caps run out of voltage. He is using the circuit to de-power his trains from their LiPO batteries when they reach a certain level.

As I reviewed that post, a whole 'nother method of accomplishing the goal presented itself in my mind. This one does work. Stay tuned.

This was the next itteration of the project. In this version we latch the central relay, but we use a resistor to do it. Then, when we want to unlatch the relay, we introduce a second resistor in parallel with the relay coil and ground. When this additional load is imposed, this becomes the “straw that breaks the camels back,” the relay no longer receives enough current to hold it open, and it shuts.

In this case, the relay on the right has to throw completely to break the circuit rather than just “twitch” off of its seat. The values are still fairly critical and introducing the load to the latched relay may require some revising to the caps and resistors.

Of course your values for caps and resistors would vary with the relays used and loads imposed.

You could use a J-K flip-flop like a 7470 and connect your momentary controller to the clock input, and set up J and K so Q toggles back and forth.

Or, and I could have generated the pulse using a 555 chip. But I was trying to keep it a “passive” circuit and not use anything I didn’t already have.

Sounds like it would be much easier to just use a stepping relay.

Dan Pierce said:

Sounds like it would be much easier to just use a stepping relay.

Takes all the fun/enginuity out of it.

“First the “ground rules.” This had to work with no more than the two, 12 volt, dpdt relays, and any capacitors and resistors I had at hand.”

And of course there is always the personal satisfaction in accomplishing something in a way that apparently no one seems to have done before, at least that I can find.

Todd bought a ton of these relays, uses them for everything. This is the untold part of the story. He does indeed use them successfully on his automated railroad.

I gave up a while ago on suggesting anything that is solid state. (http://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-tongue-out.gif)

I see it’s a personal challenge, but I shy away from anything mechanical given a choice.

This method is fine given that Todd has all these relays, uses them all the time, enjoys spending time to figure out how to use them in many different ways. And yes, from that perspective, his solutions are usually ingenious.

But nothing I would recommend to others that just want to get the job done. Todd and I have banged heads before on this and I’m sure there will be a head-butting retort. (http://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cool.gif)

Greg

I am old school. My 2 reverse loops, one has 2 stop blocks in it, and single track mainline set up, is controlled by 2 12 volt relays, using 2 5 volt relays as my current sensors. So I enjoy Todd’s use of relays, even if I have no need for such a circuit at this time.

And Greg, Dave Bodner started explaining how my railroad could be controlled with a Pic, the first time he was at my house. I prefer relays in circuits with higher current draws. Relays tend to be bullet proof, whereas solid state thingies tend to let loose their magic smoke right when it would torque me off the most. That is just my opinion, based upon my past experience.

Since I have been hitech my whole life, I like doing things with more modern compoments and when you have electronics like the revo then I choose to use over the counter open collector drivers to get the 10ma of current to drive 200ma relay coils, or incandescant lights without adding resistors.

My favorite is the 75452, 2 open collector drivers in one package.

Gee, so you are going to entice me to look through my semiconductor catalog, aren’t you? I use semiconductors (chips) when I think they will do the job. As for power polarity control on my track, I will stick with relays. I have given up on using tubes, since they are so hard to come by anymore.

With modern FETs, in small sizes that handle 60 amps and very low on resistance, many things are now possible and reliable.

Also, as the desired functions become more complex, it’s increasingly difficult to implement with more basic logic pieces.

Greg

True. And teh FETs work a little bit like the old tubes. But they too can be made to release the magic smoke, and in fact, the one I did that with in school, actually shot part of its case across the room. (http://largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-surprised.gif)

Did I mention that my nickname, when I was in tech school, was “Flash”?

Yeah, but was it a 60 amp one? If you look up the parts on modern decoders, many are now 50 and 60 amp… you really have to work hard to kill them, not as easy as it was years ago… of course years ago we wondered what 110v ac would do to TTL logic… fun to blow the lid of the package!

Greg

110 vac to a chip. Yup, that’s how I got the nickname “Flash”. But even more fun is 110 vac to an electrolytic capacitor. Point the rubber stopper toward the ceiling and let her go. After the bang, all this paper and foil confetti comes drifting back down.

It is also fun to tie the 5kv in old tv’'s with pencil lead on a bench and watch the sparks and flames!!

Yeah, a lot more fun than smashing up an old tv and learning how long a charge can last on a 27" picture tube! Bzzzzttt… Don’t ask me how I know…

Greg

Greg, I know. The worst part was having the whole class standing around saying “do it again”, while I just stood there shaking uncontrollably.

Greg Elmassian said:

I see it’s a personal challenge, but I shy away from anything mechanical given a choice.

This method is fine given that Todd has all these relays, uses them all the time, enjoys spending time to figure out how to use them in many different ways. And yes, from that perspective, his solutions are usually ingenious. (http://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cool.gif)

Greg

Who am I to argue???

BTW, the two relays are 12 volt with 280 ohm coils. The “charge/discharge” capacitor (C3) is 330 mfd. The cap (C1) on the “make” (center) relay is 100 mfd and the cap (C2) on the “break” relay to the right is 22 mfd. Use 16 volt (or more) caps.

The “latching” resistor (R1) is 12 ohms. Use at least a 1 watt resistor here because rapid, continual cycling of the system (not something you would do) will cook a 1/4 watt resistor in this position. The unlatching resistor (R2) is 2 ohms.

Greg Elmassian said:

Yeah, a lot more fun than smashing up an old tv and learning how long a charge can last on a 27" picture tube! Bzzzzttt… Don’t ask me how I know…

Greg

Boy does that bring back memories. Back in the olden days (1970’s) I had been playing with electronics for years before I finally took a class. I became known as the ‘absent teacher’. When the teacher left the room, I would show some of the neat things you could do that would give some “Dazzling Effects” that the teacher would frown upon. I was finally asked not to come to class except on Friday at the beginning of class to take the weekly test. Took all the fun out of the class.