Logic puzzle - 4017 sequencer length

[CHRISKELLY]

Been pulling my hair out trying to solve this

I've got two 4017's chained to create a 16 step sequencer as per the datasheet. Theres a 16 position rotary switch used to change the sequence length from 1-16 total steps.

Been trying to add a 3 position switch which will allow 3 states:
1. Default to 8 steps
2. Default to 16 steps
3. Sequence length set by rotary switch.

Using a double pole toggle switch (on - off - on).

It's working OK so far except there are two glitches:
1. When in '8 step' mode, if I switch to 'rotary' mode when the sequencer has already passed the step set by the rotary, then the sequencer carries on to 16 before switching to the custom sequence length set by the rotary. (i.e it misses its chance to reset and carries on until the default reset after step 16)

2. When in '16 step' mode, if I switch to '8 step' mode during steps 9-16 then the sequence instantly jumps to step 1 which must be triggered by pin 9 of 4017#1 being held high during steps 9-16. (In 8 step mode 4017#1 resets on pin 9)

What I'd ideally like is for glitch 1 - to reset after 8 steps first
And for glitch 2 - to finish the full 16 steps first

Hope any of this makes sense! Schematic attached.

Any suggestions whatsover are welcome

---

Untitled.png
Description:
Filesize:	20.5 KB
Viewed:	486 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[JovianPyx]

Does it work correctly when using only the rotary switch to select sequence length? That is, do the 8 and 16 glitches stop in rotary switch mode?

If so, then why not eliminate the 8 and 16 switches and just use the rotary?

[CHRISKELLY]

Yeh it works fine with the rotary on its own.

Maybe you're right perhaps I should just stick with what works

[PHOBoS]

ah fun another puzzle

I will look into it but at first glance I noticed a couple of things:
- pin 1 of the 4069 should have a pulldown resistor
- pin 15 (reset) of 4017 #1 should have a pulldown resistor.
- pin 5 of the 4066 is always high.
- there are a couple of unnecessary diodes.

There also seems to be something wrong with the way the 2 4066 switches are connected to pin 13 (CLKEN) of 4017 #1

[PHOBoS]

I haven't really looked into how and why it works/doesn't work but I cleaned up the schematic a bit.
I am pretty sure U3b can and rather should be discarded. It can probably also be slightly simplified with a mux.

---

16_8_custom 4017 sequencer.gif
Description:
Filesize:	67.5 KB
Viewed:	443 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[CHRISKELLY]

I'm impressed how tidy you made the schematic look Phobos!

Thankyou for doing that, and for the general sense-check.

I'll go and check my wiring then report back

[PHOBoS]

I think there might be one or more errors in the schematic you posted as I don't see how the 8/16 step modes could work.

[CHRISKELLY]

I better go check

[CHRISKELLY]

OK I think I might have it:

Pin 9 of 4017#1 sends a 1 into an AND gate. Pin 3 resets a latch (e.g on a 4027) and Pin 2 sets it again but via a 4066 (normally closed). When the latch is set HIGH it enters the AND gate to let pin 9 reset 4017#1

In rotary mode or 16 step mode, the 4066 is opened and the latch does not get set HIGH by pin 2.

This means that changing modes (e.g from 8 step to rotary steps, or from 16 steps to 8 steps) only happens after the sequencer has reset and has a chance to go past step 2 (pin 2 of 4017#1)

A sorry looking schematic is attached

---

20180821_235109.jpg
Description:
Filesize:	4.35 MB
Viewed:	409 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[CHRISKELLY]

Finally solved this!
Phobos you were right - my first schematic wouldn't have worked. I'd drawn it wrong

The reason it glitched when going from 8step to rotary modes was if the sequence was already past the rotary reset point, then the sequencer had no instruction to reset itself.

The reason the sequencer reset when changing from 16step to 8step mode was that the clock inhibit on 4017#1 was being sent low by changing to 8step mode. This started the clock again on 4017#1 which reset itself after pin 11 (after all 10 steps of the 4017 had finished) which reset the whole sequencer.

I fixed this using latches for both: set on step 1 and reset on step 2 but only when in certain modes. For this i just used a 4066 switch to either allow or prevent the latch from resetting itself.

This meant that changing modes only happens once the previous mode has run full cycle and can be set again on step 1.

Bit of a pointless post tbh but I was really pleased!
I'll do a schematic later

[CHRISKELLY]

Here's the schematic. Hope it makes sense. If anyone can spot any redundant parts please let me know.

---

4017_sequencer.pdf
Description:		Download
Filename:	4017_sequencer.pdf
Filesize:	54.06 KB
Downloaded:	406 Time(s)

---

[PHOBoS]

I only glanced at it and did spot some things but instead of redrawing it I think I will make a thread with some guidelines
for designing digital circuits. Then you can first try to clean it up yourself which I think would be more helpful.

[PHOBoS]

ok, I made a thread which might help you a bit.

http://electro-music.com/forum/topic-70288.html

[CHRISKELLY]

Thanks Phobos. Good of you to put all that info together