Running a 4017 backwards?

[adamon]

So I'm working on a few new sequencer ideas, and some could benefit from having a forward/reverse function (switch controlled). I think I've come up with a crude/brute force method of implementing this:

Basically just using two 4017s with their outputs tied together through diodes to the main outputs 1-8. One would have its outputs going sequentially 1-8, the other 8-1, so they are "reversed" from each other. Then just use a spdt on the clock input to choose which 4017 to go to (seems like there might be some possible issues with keeping the 4017 not in use "down" so it doesn't leave a step on...). A single reset jack could be wired to both of the reset inputs.

This seems like overkill, which a lot of my stuff is, so I'm just curious if anyone has any other ideas (or other chip suggestions...).

[blue hell]

There are binary up/down counters, you could use that followed by a demultiplexer like 4051/53/53 or a decoder like 74138 to get a step sequencer.

[stolenfat]

i considered this idea one time... its messy... but use two 4017s... the 'direction' switch controlling forwards or backwards would simple redirect the clock signal to a 2nd 4017, which is wired 9 to zero instead of zero to 9 (backwards).

I think each could would use the same pots with duplicate connections running from each chip... i think so at least, maybe toss some diodes on there just to be safe. But theres an idea. work it out, give it a shot


edit: ha! same idea, shoulda read your post first before i reply'd

[stolenfat]

well, here.. in regards to using two chips and the worry of it holding down a step. use a triple throw, dual pole switch?? which would control the direction of the clock signal and also cut the power to the 4017 not being used while connecting the power to the other. this might solve your problem.

[Uncle Krunkus]

I agree with Jan.
Use a binary up/down counter, a second chip to "reverse" the count, and a binary to decade decoder to get back to 1 of 10 outputs.

[adamon]

thanks for the ideas! I've got some demux's on my next order, so I'll try and play around with using one as bluehell mentioned. I haven't messed with decoders much yet, but I'm sure some reading will be enlightening.

With a up/down counter, would I still need a second, "reversed" one as well? Shouldn't the up/down part take care of this?

stolenfat: yeah, using a larger switch was an option I considered, but I hate how much switches cost! I've already implemented some similar chip-power switching to holding things down, but it just seems like overkill at this point (that and I think that it's time for me to start moving beyond the basic 4000 series chips that I've beaten to death... almost... ; ) )

Thanks for the feedback as always!

[blue hell]

[PHOBoS]

As stated above, a binary up/down counter will do the trick. Since I don't see one mentioned, let me make a suggestion: 4029 (followed by a 4051).

connect the four parallel data inputs (P0,P1,P2,P3) to ground and use the parallel load pin (PL) as a reset.

[adamon]

So I finally got around to wiring this up today, and I'm having some trouble; it doesn't seem to be sequencing in the correct order. I've got a 4029 counting in binary (correctly) that I'm feeding into a 4051, and I've got the outputs wired in the order that should be 1-8, but its not going in order. I could just figure out the order its going in and use that new order re-arranged to go 1-8, but it seems like I ought to be able to use it like it is intended to be used.

I've got pins 6 and 7 (inhibit and low input) on the 4051 grounded; am I missing something else that needs to be high/low?

[adamon]

Yet another "doh!!!" moment for me... I had the 4051's a,b,c inputs backwards (I was going c, b, a). Sorry to be answering my own questions...

However, one issue still remains that I forgot to mention earlier: how do I get a reset function with the 4029? Reading over the link from PHOBoS, it seems like maybe I should be using a 4516 instead... not sure if I understand what PHOBoS is referring to as the "parallel load pin" for a 4029.

[blue hell]

Parallel load would be PE (Preset Enable) what is does is to copy the JAM inputs to the counter to set its start value, so when the JAMs are at zero it will reset to zero.

[DGTom]

Even better. Wire the JAM inputs to switches & the Load pin to a socket (via a switch) & you can 'skip' steps.

Even betterer. Wire the JAM inputs to sockets (via switches) as well & you have a sequencer that can change directions, stop / start & jump around

[adamon]

Thanks all, its working well now.

DGTom, already on that stuff! I've got on-off-on spdt switches for on/skip/reset, and I've got the direction pin wired up to a jack for "voltage controlled direction" control. Endless possibilities! I'll be sure to get it up on the build thread with my other current projects once they're all done.

[Rykhaard]

Deleted

[adamon]

Hey Rykhaard, its an honor to hear praise from you (as from many of the others that I've learned so much from over the past year). I'm a big fan of yours, and I thoroughly enjoy your videos!

So I've made a post about my method of implementing control signals in my system here: http://electro-music.com/forum/topic-42914.html as its a little different from the norm. Hopefully we can figure out a way to translate how my stuff works over to something that will work for you.

So from here on, I'm going to assume you understand how I use "resistance" control rather than direct voltage control.

The meat and potatoes of the sequencer is what is described above; I've got a 4029 driving a 4051 to produce the sequence. The common pin on the 4051 (pin 3 on the chip i think) is tied to V+ (9V in my case) and each of the outputs is tied to ground. Each output channel (the sequence steps) has two lines. The outputs each have a resistor and led line to ground to serve as the step indicator (the yellow leds within the circle). Each output channel also has a separate line which are all summed to a single resistor, pot (used as "amount/offset"), and led which happens to be half of the output vactrol. These summed lines have the individual step level pots (the circle of knobs) before summing, so whatever the pot is set at, it will drive the output led at that level (plus whatever the output "amount/offset" is set at). The on/skip/reset functionality comes from simply using an on/off/on spdt switch at the beginning of each step output from the 4051; on = step is on, off = skip since nothing gets through to the output led, so nothing happens, and the other on is wired to the reset bus, so whenever a step is set to reset, it will simply trigger the reset input on the 4029. I also wanted to have direction control over the sequence, hence I used the 4029.

So it seems like you should be able to just make a normal cv output by just getting rid of the vactrol and wiring a jack in instead (to the signal and to ground). The indicator leds could still simply be tied to the modules ground. Should that work?

[adamon]

Wait, shit... maybe I'm labeling my functions wrong; by skip to you mean actually skip the step, like loose a count in the sequence, or have a step that is simply off... I assume I'm calling this the wrong thing... sorry if I am.

[Rykhaard]

Deleted

[DGTom]

Sounds like you have something brewing Ryk

The "skip" I was talking about isn't a proper 960 style, per step skip either.

Using the JAM inputs & preset enable control on the 4516 I'm able to patch in a single skip, here is the schem for the device I use it in;

http://dl.dropbox.com/u/252415/PSPS_.png

This is designed for analog modular use, so it has comparators on the inputs & transistored buffered outputs. It has a decoded 1 - 8 outputs, so, by patching out 4 into the Preset Enable input & setting the preset switches to, say, step 6, I have a counter which will go 1, 2, 3, 6, 7, 8.

So really you only get 1 skip per cycle so to speak. The cool thing is with this the binary outs also skip, so you get an odd count & divisions. I find it most usefull for drum seq'ing, especially when used with external control of the direction of the count.

In the above counting example you can add an external toggle flip flop. Connect step 8 to Set. So you get counts something like; 1, 2, 3, 6, 7, 8, 7, 6, 5, 6, 5, 6 & so on, until you reset the counter and / or flip flop.

Great fun

[Rykhaard]

Deleted

[Ajax]

Is there a difference between a 4017 sequencer and the Baby 10? I'm trying to find a non-CV sequencer...

[-minus-]

The Baby 10 utilizes the 4017 IC. Typically it consists of a 4017 clocked by a 555 timer. What purpose was the sequencer to be used for?

I recently built a sequencer based on this but wanted to use 8 steps not 10. Also I wanted it to trigger drum modules and be capable of sending out a trigger pulse on consecutive steps. I did this by inverting the clock signal and using NAND IC's. The step output of the 4017 and the inverted clock are fed into the 4011 configured to AND.

[tjookum]

[-minus-]

Ah, yes... I did see that on Deathlehem... I found that to make consecutive steps trigger, the inversion of the clock signal was important. Need a really short pulse!

From this:
____ ____ ____
_l l_l l_l l_

To this:
_ _ _ _
l____l l____l l____l

But hey Tjookum... you're a General! You already knew that. I just always wanted to draw one of those diagrams using just the keyboard to draw with... You know, like those people on this forum who ACTUALLY know what they are doing!

EDIT! Oh what a *0$@** laugh! The diagram didn't work!!! It came out all f'd up!! Ha ha! I could fiddle with it, but I wont! It's far more amusing as it is!!!

From this:

To this: