from an email:

I am curious if there is an easy modification I could do to these sequencers so that it would expand the steps by "daisychaining" them? Trigger one from the previous etc. in a cycle. I could figure out how to patch this "event" using my modular but that would just tie up important modules.
Thanks,
- michael

Chaining the sequencer’s together would require some non-trivial circuit mods. The 4017 decade counter was chosen because of how easy it was to use and how cheap it’d be to produce the kits. To chain 2 together you have several options. First, you can make 3 seperate sequencers. One of these will control which of the other two is activated. Shorten the steps of the control seq. to 2 by inserting a patch from the 3rd step’s gate output to the reset jack. This will play a 2 step sequence. To clock this control sequencer, you need the signal from the last step of each of the other 2 seq’s. This means that when one seq finishes, it will send a signal to the control seq to advance a step. Now, the last part (note, I haven’t tried all this myself yet), connect the clocks of the 2 sequencers together. slaving one to the other. You will need to modifiy the clock on/off signal to allow it to be patch in externally. Then connect the output of step 1 of the control seq to the clock enable of the first seq. Similarly connect the output of step2 of the control seq to the clock of the second seq.

That’s kinda messy, but it should work. The cleaner option would be to alter the internal circuitry. There is information on chaining 4017s on page 5 the datasheet:

http://focus.ti.com/lit/ds/symlink/cd4017b.pdf

You would need to add a CD4001B to get the NOR gates necessary to make this work. not very difficult, but you’d probably have to cut a few traces on our circuit board.

Hope this helps.

-Travis

More options…

The trouble with chaining two of them together is that even if you could get one to pause while the other was moving, it would still have an output on. To get around this, you’d need to either use a transistor to cut the power to the whole chip or not use all the outputs, as described in this great pdf.

To summarize how it works, the 10th step of the first 4017 is connected to its own clock inhibit input. So when the 10th step is reached, it stops. The 10ths step is also connected to a NAND gate that turns off a clock inhibit on the 2nd 4017, which has been pausing in the 1st position (not connected to the output). So the 10th step actually comes from the 2nd position on the 2nd 4017. When the 2nd 4017 reaches its 10th step, it connects to the same nand gate to turn itself off, and also to the rest on the first 4017.

You’d need a bunch of diodes or some switches as shown here to combine the signal in the end. Also, you’d have to figure out a way to reset all the chips when everything is first turned on since they don’t start in a known state.

The spec sheet for the 4017 only shows how to hook them up so that the 2nd 4017 steps every 10 counts of the first…

-scott

lose q8 and q9. we now have an 8 step sequencer.

invert q9 with a transistor and attach it to the hold pin.
We now have a sequencer that plays once and suspends.

patch q8 to reset to get 8 stages looping.

Patch q8 to a second sequencer reset. the first will suspend and the second will begin.
patch q8 from the second sequencer to reset on the first..

you now have a 16 step sequencer.

one problem. Because q9 stays high for a tick of the clock instead of reseting the sequencer will stay on the first stage twice as long as it should.

solution.

connect q9 to reset through a small capacitor so al it gives is a tiny pulse.

There you go. Trivial mod using 2 components… oh and maybe some resistor.

and it works:

http://www.youtube.com/watch?v=i_Z0A-qA9jA

oops I should say:

“oone problem. Because “q8” stays high for a tick of the clock instead of reseting the sequencer will stay on the first stage twice as long as it should.

solution.

connect “q8” to reset through a small capacitor so al it gives is a tiny pulse”.