4016 flip-flop and switch - all in the same chip

[bioroids]

Hi

I posted a schematic of gez paton's idea: using 2 sections of the 4016 for the flip flop and the other to for switching wet/dry signals. you can find it here:

http://ar.geocities.com/bioroids/articles/flipflop.htm

I must say it works pretty well, but there's still some clicking I couldnt get rid off. Would be nice if anyone breadboards this to tweak it a little more.

By the way, replacing the 4016 with 4066 results again in increased popping, but also would be nice to see what happens between diferent brand ic's.

Luck

Miguel

[gez]

Thanks for posting this Miguel.  It’s changed quite a bit, but the basic principal is still there.

The original schematic was for a Boss-style Bypass and used two switches in the 4066/4016 to do the honours.  I had a tri-colour common anode LED connected up as a ‘standby/on’ indicator.  The anode connection went to 9V, one cathode was wired in series with R1 and the other cathode in series with R2 (though they were 3k6 not 10k).   R4 was 470k and C1 was .47u (though I appreciate that these values will probably need tweaking for the type of switch used).

Here’s the original thread with the details:

http://diystompboxes.com/sboxforum/viewtopic.php?t=17905&highlight=

If anyone ever finds a fix for the popping, be sure to post it!

Thanks again Miguel.

[smoguzbenjamin]

Ok, please forgive my dumbness tonight but where do you actually connect the effect in/out and jacks?

[gez]

Ok, please forgive my dumbness tonight but where do you actually connect the effect in/out and jacks?

With the version posted, you don't - it's not for bypass, it's just a simple switch with LED indicator, triggered by a momentary switch.  

For bypass, the switch used for the LED in Miguel's schematic needs to be roped in to do some of the work.  One switch comes off a input buffer the other comes off the FX output, both are connected and a output taken from the join and fed to an output buffer.  One switch's enable connection is triggered by one of the outputs of the flip-flop, the other switch's enable connection from the other output.  Using a common anode tricolur LED in the flip-flop section gives you a 'status' indicator.

It's a variation of what's normally done with FETs in the bypass used in many commercial FX units (well, older ones).

[bioroids]

Hi

Yes I modified quite a bit. I posted the final circuit I had tested with no (or little) pops. The idea was to share it and add some more minds tweaking it into a fully functional switch.

I should post a "complete" switch scheme anyway so it's ready to be used as is.

Luck

Miguel

[bioroids]

Hi!

I just posted a modified version on the prior schematic. This time is a complete switch, only lacks the input and output buffer (you can use anything you like for that: opamp, fet, bipolar..).

http://ar.geocities.com/bioroids/articles/flipflop.htm

I hope it's clearer now.

Luck

Miguel

PS: I just added a few samples from my cloned Pulsar Tremolo, just for the sake of it.

http://ar.geocities.com/bioroids

[gez]

Thanks for posting!  Hopefully someone might suggest a fix for the clicking.

[gez]

PS.  If you take out C4 & C6 and use only one of them coming off the ‘output’ lead, you only need one resistor to do the biasing honours (get rid of either R6 OR R8)  

Lower parts count…

Edit: that's supposed to be R8 'close bracket' but it comes out as R8)

 :roll:

[bioroids]

Hi there

I keep discovering trouble with this

There's some trouble with the RC network to slow down the control voltage into SW3 and SW4. If it's too slow, the different discharge/charge times lead to uneven switching of the wet/dry signals (very ugly). If it's too fast, althought the switch itself doesn't click, the abrupt change of the signals introduces again the popping. Ahh, there has to be a solution anyway.

I think an aproach could be make more even the charge and discharge time. The trouble is the great impedance of the control inputs on the chip, so maybe putting a resistor to GND there could help.

But that introduces the issue of a reduced HI voltage (because of the voltage divider we got now), so we may need to introduce an aditional stage here. A transistor should be enough. But that defeats the benefits of having just one chip for all the switching.
:?

Things to keep trying, If anyone has an idea, please post it!

[smoguzbenjamin]

Wow. I'm really impressed, man! Whenever I get a dud Boss pedal I can now replace the switching! When I get enough money to play around with I'll definately try this out!

[gez]

Just looked through an old Penfold schematic (chopper tremolo type circuit) which uses the 4016.  Instead of using resistors at each end of the switch going to half the power supply (divider), he directly couples the output of an op-amp input buffer to the input of the switch and directly couples the output of the switch to another op-amp follower via a 10k resistor.  There’s a 10n cap from the junction of the 10k resistor and the switch’s output to ground.  The output buffer has a MOSFET input (very important!)

Here’s an edited version of the blurb:

“The output buffer amplifier constitutes a basic sample and hold circuit.  C4 (the 10n capacitor) has no significant affect on the circuit when the electronic switch is closed as it is driven from a fairly low source impedance.  When the switch opens, C4 holds whatever signal voltage happened to be present at that instant.  The ultra-high input impedance of IC3 ensures that C4 does not discharge significantly while the switch remains open.  The charge on C4 immediately jumps to the new signal voltage when the electronic switch closes again.  This helps minimise switching clicks, whereas simply leaving the input of IC3 (output buffer) floating during the ‘off’ periods would almost certainly exacerbate the problem.”

The answer might be in how the buffers are connected then, thoughts anyone?  I realise the above description is for a different application, but perhaps the sample and hold idea would prevent voltage differences at both ends of the chip? (edit: I meant switch, though in the cold light of day I haven't a bloody clue what I meant!)  In a bypass (if the FX output used a op-amp too) one amp's output would always bias the final output buffer.  

It's late, I'm talking gibberish! :roll:

Another thing of interest in the blurb are his comments on the 4066:

“The 4066BE is not recommended for use in an audio application such as this.  The 4066BE is pin for pin compatible with the 4016BE, and tends to be regarded as a superior version of the 4016BE having a lower “on resistance”.  However, the 4066BE is not totally compatible with the 4016BE, and it tends to produce switching ‘clicks’ when used in audio applications”.

[rx5]

I think m gonna bread board those again.... tyr to eliminate the "pops"......


-Regards


Ralph

[gez]

Whenever I get a dud Boss pedal I can now replace the switching! When I get enough money to play around with I'll definately try this out!

We need to iron out the problem of clicking first!  That aside, if you get dud switching it's easier to try and sort out what's already there rather than squeeze in a new switching circuit.

[smoguzbenjamin]

Yeah true, but I was thinking more along the lines of "Here have this dead pedal I don't use" and I have no use for the electronics inside but only for the box. Something like that

[bioroids]

Hi there, I'm downloading this thread to read at home with a fresh mind.

Meanwhile I posted a minor change in the schem, as the prior schematic didn't happen to work (because of the way the led was hooked).

Luck

Miguel

[gez]

Hi Miguel

It’s probably an idea to make R2 (10k) smaller to get the LED to light up reasonably well.

I used a common anode tri-colour LED (which is basically two LEDs in one), but you could change the LED shown in your schematic to green and wire a red one to the other switch to give you a ‘standby/on’ indicator (might as well use the current that flows through the other switch when on).

[gez]

PS.  Penfold seems to have got 'open' and 'closed' the wrong way round in his blurb.  

I think the idea behind his circuit is that the small cap (10n) 'absorbs' any current which flows when the switch opens, thereby preventing clicks.  Because it's such a small value and the switch is driven by a low output impedance buffer, it doesn't slice off all the high end.