4066 switch bypass layout

[Rob Strand]

Rather than using the 4013, we'd have to use a NAND chip and roll our own flip-flop (so we won't get two out of the one package)

At first I thought that's it but then I realized when you roll your own it doesn't have a clock input, only R & S.

The more I think about it that two-gate toggle circuit is looking like a good choice.  Only two inverters (or Nands).  The downside is what to do with the rest of the chip.

DOD's CMOS 4007 based circuit was another one.

Switch:
CD4066: four individual bilateral On/Off switches, many different possibilities.
CD4053: three individual unidirectional SPDT switches, easy 3PDT wiring.

Perhaps a JFET option for the single switch case like Delays, Flangers and Chorus (if you want to just switch the delay path off).

I am planning to make a modular design

As a separate board or as a re-useable circuit block?

[reddesert]

Here is an OSH Park PCB layout for the Wicked Switch described by tonegod, made by alanp, who is a user on the madbean forum and maybe here. https://oshpark.com/shared_projects/wOjwV7Vu

[Fancy Lime]

The more I think about it that two-gate toggle circuit is looking like a good choice.  Only two inverters (or Nands).  The downside is what to do with the rest of the chip.

Yes, me too. For controlling a 4053 we only need two inverter or NAND gates. For a 4066 we need Q and -Q outputs so we need one extra gate. Not using all gates of a chip doesn't really bother me all that much. We can use one extra gate to drive the status LED. If we use a Hexinverter, we can for example drive two 3PDT switches plus two LEDs using two 4053's or two bilateral DPDT switches using two 4066's (in which case we have to either power the LED's with separate transistors or see if we can get away with taking the LED current right off Q without an additional buffer). A Quad NAND could drive one 4053 or 4066 plus LED.

Perhaps a JFET option for the single switch case like Delays, Flangers and Chorus (if you want to just switch the delay path off).

Great idea! We can probably drive an almost arbitrary number of JFET switches either a s On/Off switches of functions or (when feeding some Q and some -Q) alternating switches to switch signal paths and whatnot.

As a separate board or as a re-useable circuit block?

Yes. I mean: both. In fact several separate pad per hole boards for all different control and switching circuits. Depending on the application and what chips you want to use you can combine those to taste. Like a switcher buffet! I will try to make it so that the same layout works with strip board as well for maximum re-usability when pasting the circuit block into a larger board. I'll have to see if that works, though. My preliminary sketches need intolerably complicated traces on strip.

Andy

[Rob Strand]

If we use a Hexinverter, we can for example drive two 3PDT switches plus two LEDs using two 4053's or two bilateral DPDT switches using two 4066's (in which case we have to either power the LED's with separate transistors or see if we can get away with taking the LED current right off Q without an additional buffer). A Quad NAND could drive one 4053 or 4066 plus LED.

Should come together nicely.    I'm all for using the free gates to drive the LEDs.   These days the LED current is low enough not to cause problems.

Yes. I mean: both. In fact several separate pad per hole boards for all different control and switching circuits. Depending on the application and what chips you want to use you can combine those to taste. Like a switcher buffet!

OK.  The underlying structure is good enough to do all that.

[ElectricDruid]

I like where you're heading with this. Versatile and simple. Nice work.

Tom

[Rob Strand]

FYI:  I came across this 4013 based toggle circuit.  I found it by chance a while back but couldn't remember how I got there.   It's the second and third circuits on this page.

http://www.pcsilencioso.com/cpemma/flipflop.html

The advantage of this circuit over the Dan-Electro circuit is the slow rising voltage is disconnected from the 4013 clock line.   The low value of R2 dumps all the charge from C1 locking out bounce.

It came from Spehro Pefhany on this post,
https://groups.google.com/forum/#!topic/sci.electronics.design/I7hA2tEhj8o

I used to yack with those guys on sci.electronics.design about 20 years ago.