40106 Tremolo - Seperate grounds?

[mr.adambeck]

Hi, I recently have been going through the projects in Nicolas Collins' "Handmade Electronic Music: The Art of Hardware Hacking."  It's a really, really fun and informative book for noobs such as myself!  One of the projects inside is a very simple tremolo based off of the oscillators of a Hex Schmidtt Trigger (in my case, I'm using a 40106):



According to the book, you don't connect the ground of the circuit to the ground of the audio as "this is important for maximizing the audio quality of this circuit!"  Which, if I were planning to run this off of a battery forever probably wouldn't be a problem, but since I plan on using a variation of this (a multiple LFO tremolo) as a stompbox on my pedal board a different issue arises, since the ground of the power supply is connected to the ground of the audio in the majority of my boxes.  Wouldn't this mean if I connect this pedal to the same power supply the ground separation would be broken?

So my questions are: a) why does this circuit need to have the grounds separate?  b) am I correct in my thinking that connecting it to my power supply would then connect the circuit's grounds to the audio grounds? and c) if this is the case is there any way of preventing it?

Also, just to clarify, the left LED in the schematic is just an indicator LED so you can see the tempo.

[glops]

I am very sorry to report that I do not have an answer but was reading that very same topic today from the same book on my flight
back home.  I want to build that circuit, as well, and if I find any info or figure anything out I will report back...

[mr.adambeck]

glops--  That's funny!  Small world I guess!  It sounds great as done in the book, and I found it especially fun to get two or more oscillators going for really weird patterns (the stranger the effect, the better, as far as I'm concerned).  Let me know if you find anything out!

[JKowalski]

I believe it is to alleviate ticking problems (the square wave transitions of the LFO being heard audibly through your amplifier)

Yes if you have a daisy chained (non isolated) multi power supply it would connect to audio ground through it.

Use a different power supply?

I think that it may not be that bad of a problem, but I can't guarantee it. As he said, it's not necessary but may "maximize the audio quality"

If you can't do it at all, and you have problems as above, you can put a capacitor on the output of the LFO to slow down the square wave transistions...



Just a note, if you are using two different type LEDs there they might mess with each other - one might "steal" the others current, as the amount of current both LEDs can get together is limited by the 1k resistor and the voltage output of the 40106...

If one LED has a smaller voltage drop than the other one, the voltage on the anodes of the LEDs will both be dropped down to that voltage... And that means that the one that needs the HIGHER voltage drop to turn on can't anymore.

Only important if you are using different type LEDs, or the same LEDs but with widely varying specs. This whole issue can be solved by giving each one it's own 1k resistor instead of them both sharing the same one - if it ever even becomes an issue...

[glops]

Thanks for that info, JK.

I got the book for Christmas and it's been the best read so far for me concerning chips.   I actually understand some things now because of the
explanations with the illustrations and pics included.  I think by completing some of those exercises I will learn a lot and be able to understand
some more of the technical dialogue here on this forum that is still a foreign language to me.  Just got back from Texas and plan to dig into
some of the exercises tomorrow and will report back!

[mr.adambeck]

I built a Escobedo PWM (which is also based off of a 40106 chip) with an LFO that had ticking issues and putting an input buffer on the circuit seemed to get rid of the ticking.  Would that work in this instance?

[JKowalski]

I built a Escobedo PWM (which is also based off of a 40106 chip) with an LFO that had ticking issues and putting an input buffer on the circuit seemed to get rid of the ticking.  Would that work in this instance?

I think that the reason that worked is because the buffer lowered the impedance of your signal - low impedance signals have a harder time picking up unwanted noise...

It might work.

[mr.adambeck]

Thanks for all the advice J!

[mr.adambeck]

Another question:
If I just put this after a pedal with a buffered out (IE my Boss Tuner or Boss Chorus), would that work the same as building an input buffer on it?  This would make the design of the pedal so much simpler!

[earthtonesaudio]

Again maybe.  If the cable between the two is reasonably short, the signal impedance going in is probably still going to be low enough to shrug off ticking from the LFO.  Again, that's assuming the ticking is coupling into the audio path via that route.  If it's coming from somewhere else, neither an internal or external buffer will help.

[glops]

Mr. AB,

I got around to building a couple of the exercises from the book.  The trem from the pic you posted works but I think I used the same ground on my breadboard.  It sounds good but needs tweaking.  I am getting the clock ticking but it doesn't bother me at this point because it's my first tremolo.  Did you get any of the circuits going?

[mr.adambeck]

I've breadboarded a lot of the synths.  I even breadboarded the trem but didn't mess around with the ground. They've all worked great!

[Top Top]

That is such a brilliantly simple tremolo. I'll have to try it out on the breadboard.

[glops]

It's probably the easiest thing I've built on a breadboard.   I have been doing fuzzes and was looking for a tremolo but was intimidated by the schematics but this gave an immediate sound and hopefully with some tweaking it will be better.  I ran some distortion through the circuit and it sounds awesome.  That book has been a tremendous help in getting me into chips....