4017 sequential switching with a momentary switch

[deadastronaut]

looking for a schematic that does this...everyone has vids but no schemo attached

https://www.youtube.com/watch?v=7llY8nexTvI

or this

https://www.youtube.com/watch?v=dOMaGSt6Rqo

basically i would like to use it to replace several spst switches (up to 10 is possible on the 4017)....

how i go about this is another matter...but i just thought it might be neat to use something like this rather than a big old clumpy rotary switch...

i have a 4017 and a few momentary switches..

ideas?..simpler the better....cheers guys.

[Mark Hammer]

http://hammer.ampage.org/files/quadsequentialswitch.pdf

[deadastronaut]

brilliant , cheers mark,

looks a little more involved than i initially thought, but nice.

could i use any opamp?..

[R.G.]

That's a good one!

If you are looking for both up- and down-sequencing in a better analog of a rotary switch, there are up and down counters. I don't think they have internally decoded outputs, though, so you'd have to follow the counter with a decoder chip to make the one-of-N output. The clocks on these come in both logical flavors - an up/down signal plus a clock that says "change the way the u/d pin says to", and two clocks, an up clock and a down clock.  

If you notice, the schemo shows an opamp cleaning up the clock signal for the CMOS. Most of the CMOS counters need a clean, single transition clock signal, so there's generally another chip involved there.

Before I started messing with uCs, I'd have been happy there. But once you're over the programming hump, uCs give you enormous flexibility. For this application, today I'd just pick a $1.50 PIC with enough I/O pins, write up the BASIC program to make it read up and down clocks, and put out a decoded output on the rest of the pins. $1.50 is remarkably close to the cost of two CMOS chips, an opamp, and assorted resistors and caps, and is a single DWIW (Do What I Want) chip.

[samhay]

This is another example where a microcontroller would make the job a bit easier...

Edit - RG beat me to it.

The only thing to add is that you need to add a 60p voltage regulator to your BOM if you want to play with a uC.

[armdnrdy]

I agree with R.G. and Sam but...

I wouldn't even know where to start writing code to program a PIC. I know that one day I'll have to get over that hump due to the sensibility factor! 

In the mean time....I came up with this switcher. (for the code writing handicapped) I cobbled together ideas from R.G. and other things I found on the net. I had this circuit on the bread board for some time fine tuning it.



This circuit allows the user to choose between eight channels moving up or moving down.
You can move from channel to channel with each press of the up or down momentary switch.

There is a nice feature that I worked out which allows the user to keep either the up or down switch depressed and the channels will "scroll" through until you let off of the switch. Not unlike a T.V. remote.

I worked out the scroll feature so that the channels move at a usable speed.

[jubal81]

Also coding-impaired. This is what I came up with to cycle through five effects and it works great.

Switched debounced and power-on reset.

EDIT: R9 should be 220K - Not 220R

[R.G.]

Nice one. It's a good solution.

There is a sneaky, dirty trick that can simplify that for you. In the CMOS Cookbook, there's an example of a click-then-repeat oscillator made from one Schmitt trigger gate. I think it would do one click when pressed, then do repeats after some delay, much like (typing) keyboards do auto-repeat if you hold a key down.

[armdnrdy]

Nice one. It's a good solution.

There is a sneaky, dirty trick that can simplify that for you. In the CMOS Cookbook, there's an example of a click-then-repeat oscillator made from one Schmitt trigger gate. I think it would do one click when pressed, then do repeats after some delay, much like (typing) keyboards do auto-repeat if you hold a key down.

I'll have to take a look at that one. Simpler and less components = better! 

As it stands now...an additional 4051 has to be added in parallel with IC3 outputs (QA-QD) and a BJT driver and LED per channel has to be added for LED channel indication.

So this circuit is fairly component heavy for what it accomplishes.

[deadastronaut]

awesome work guys,

yep i'm one of the code impaired noobies for sure...these options are great.

and yes, less components the better is up my street too..

[armdnrdy]

I checked through all of my reference material files and can't find the cookbook! I thought I had it.  

So I just purchased a used copy from Amazon.

In the meantime....if anyone has the circuit that R.G. is referring to, and could post it, I would appreciate it.

It's for the good of all mankind!  

[duck_arse]

I, also code impaired, had a go of a sort, also 4017 impaired. didn't get any replies though, maybe it's the black background.

http://www.diystompboxes.com/smfforum/index.php?topic=106502.0

[PRR]

[R.G.]

That's it. I haven't broken out either the scanner or the drawing program today. Thanks, Paul.

Note that changing the R and C changes the timing.

[PRR]

> changing the R and C changes the timing.

I suspect the 3:1 ratio is "built in" by CMOS Schmitt voltage levels. So you could get 0.5sec delay and 6/sec repeat, but not a long delay and a rapid repeat or vice versa.

Also note the output is inverse of input.

[R.G.]

Yes, it is. No suspicion needed. For the typical Schmitts used in CMOS inputs, the thresholds enforce the 3:1-ish. The R and C change the timing, but not the ratios, which are embedded in the gate design and switch thresholds.

Also, the circuit depends on using the second input for the enable. Doing the same trick with inverter gates is much more complex. It really needs the 4093

The polarity of the output matters for level sensitive clocks, but not edge triggered, as each alternation gives you one edge. But if you only need one or two of these in one setup, you've bought four gates in the 4093, the others can invert the output.

It's a one-trick pony - but if that's the trick you're doing, you only need one. 

[armdnrdy]

Thanks Paul for posting the image and to you and R.G. for the explanation.

This looks too simple!

Comparing the two circuits...my circuit screams over engineered! and inexperienced!

[armdnrdy]

Here's the new and improved version with generic values.

I have to source a few 4093s, bread board the circuit, and fine tune the values.

With these values the astable multivibrator sections (IC1B, IC1D) are "set" for about 3 pulses per second. That might be a little to fast to switch between channels. I was thinking about 2 pulses per second. This will give you enough time to lift your foot from the switch to land on a target channel.

[mth5044]

Cool thread! I was wondering last week if a 'digital rotary' type thing existed. I guess this is it. It's too bad you have to scroll through all of them to get to the output that you want. I'd imagine if you were using a ucontroller, you could send out a quick reset signal then the appropriate clock to select whatever output you wanted.

[armdnrdy]

As R.G. stated...this is kind of what it is. It's not the best solution...but it is a solution. 

You have eight choices here, and depending on which channel you are at, moving up or down to a target channel only takes a few quick foot taps.

I found a few 4093s in my stash, so I'll put something more definitive up over the weekend. Maybe a video of the bread board fun.