"Stutter" Circuit

[MRTelec]

Hi,

I'd like to build a circuit that will mute and unmute the signal with variable speeds.  Where should I start?  Any help will be greatly appreciated.

Thanks.

[ExpAnonColin]

Funny, I was just drawing up a schematic for this sort of thing.  Sounds very much like a square-wave LFO modulating the amplitude... except the "positive" side of the square wave would be much smaller than the "negative" side, if you get my drift.

I have an idea, I'm going to try to work on it sometime...

-Colin

[C Bradley]

I built a tremolo that uses a 555 timer for a square wave LFO, but I used an LED/LDR setup which sounds like a blackface Fender's tremolo, not really a stutter. I'm thinking that you could use a switching transistor like a 2N2222 inplace of the emitter resistor on a 2N3904 to modulate the signal. The 2N2222 would work like a short when the square wave was at it's maximum, and like an open circuit when the wave was at it's minimum.

Chris B

[R.G.]

Pretty simple. Any way to mute and unmute works.

You have your choice of series switching or shunt switching. Shunt switching is by far quieter if you can live with the special cases, and stuttering is one of them.

Probably the simplest ways are with a JFET or a CMOS switch. I'd probably go JFET with this. An NPN bipolar can also work, but they're trickier to get right.

For the JFET:
Create a source of bias voltage. In this one case, I would connect a silicon diode anode to +9V, another diode's anode to the first cathode, then a 10K resistor to ground.  Bypass the 10K with a 10uF cap. Run a 1M resistor from this point to the JFETs drain, and a wire to the JFET's source. Connect a 1M resistor to ground, and the free end of the resistor to a 0.1uF cap. The free end of the cap connects to a 10K resistor to the drain of the JFET. Another 0.1uF cap connects to a second 1M resistor to ground.

The signal input is the 1M/0.1uF/10K resistor leading into the JFET drain. The output is the 0.1uF/1M from the drain.

Connect the gate to the anode of a silicon diode, then to a 1nF cap to ground. Drive the diode/1nF cap with a 0V to 9V square wave through a 100K resistor. The value of that resistor determines how fast the JFET turns on and off.   A 555 astable does this nicely, even though you get more control and better variation with a CMOS hex inverter astable.

9V is "signal off" as the JFET is on and shunts signal into the bias voltage. 0V is "signal on" as the JFET is cut off and the signal comes through.

You can use a lot of JFETs for this: 2n5457, 2sk30, 2n5485, j201, lots of them because the available swing on the gate is so large.

You can do the same trick with a CMOS switch chip: bias the input and output to half the chip's supply voltage,  then use it to either let signal through as a series switch or shunt it to ground like with the JFET. The CMOS will have a transient at each transition, so it will sound harsher than the JFET version.

I can draw schematics if that's not enough.

Very complicated OTA and VCA circuits can also be used, but there's virtue in the simplicity of a JFET.

[bwanasonic]

My favorite circuit of this type is the pickup selector on a 2 pickup guitar with seperate volume controls, aka Les Paul, SG etc.  That or the Fulltone Supa-Trem. As part of of a full range tremolo, it's a nice effect. If *stutter*  was the only thing  the box did, it might not be worth the jacks and enclosure .

Kerry M

[Jered]

RG on the fly. Incredible!   Jered

[gez]

R Penfold has just such a profect in one of his books (sorry, can't remember which one).  I think it's called the 'pseudo echo unit'.

It's basically a envelope follower driving a Schmitt trigger which in turn triggers a 555 oscillator.  The LFO controls a CMOS switch.  It synchs the LFO to the initial attack of your guitar so that you always hear the first note/chord, which is more musical that the random thing.

I drew up half the schematic (it's just the side chain though) for someone last year.  If you, or anyone else, want it, just leave me a message with your details (my junk mail filter seems to delete emails from this forum)

[puretube]

doesn`t "Marcos/MrMunky" have that "echo" on his site?

[gez]

It's a different circuit to Penfold's, but the one at Marcos' site is supposed to be a good choppy tremolo.  Haven't tried it mind, so perhaps it does the stutter thing too?  

Good suggestion though (wouldn't take long to breadboard and might be just the ticket!)

[Mark Hammer]

I seem to recall someone suggesting an inline mercury switch for this very task some 20 years ago.  Just shake the guitar and as the mercury rolls back and forth the signal path is disrupted and connected.  Dead simple and no batteries!  Wire up a SPST toggle to bypass the mercury switch and away you go.

Of course if you wanted to play while achieving this effect, rather than holding notes and shaking your instrument, you're out of luck.

[Arn C.]

I installed a normally closed momentary switch on a bass for my son for this purpose.  The speed is determined by how fast you can push the switch!!!!  He loves it!!!!
Peace!
Arn C.

[sirkut]

Wouldn't a rigged up volume pedal do the trick? Say have a photocell/led going inline with volume pedal, use an led flasher(lm3909) or some other circuit to flash an led on the ldr?

[Mark Hammer]

I'd counsel against going the LDR route.  While their slowness to respond is ideally suited to other functions, they are too slow to be "choppy".  Like RG, I'd stick with ther CMOS switch or JFET route.

One thing you might want to consider is using something like a 4017 or other counter for a rhythmic chopper effect.  In other words, it doesn't just go on/off, but could be "programmed" to skip beats or even emphasize beats.  Not all that difficult to do actually.

Take a look at the quadsequentialswitch.pdf document at http://hammer.ampage.org/files for some ideas on how it might be done.  Add a 555-based clock to that and away you go.  

No reason why Zachary Vex should be the only guy with sequencer-based doo-dads. :wink:

[MRTelec]

Thank you very much for your help.  I'll let you know which route I choose and how it goes.

[R.G.]

This might be a good place to fold in the pseudorandom LFO, too. The three-schmitt LFO from that article would give some randomness to stutter but with an underlying repititiveness. The three-schmitt was the basis of PAIA's surf simulator, so you can get some mental picture of the variation. You'd want a comparator so that it was only on when over "x" voltage to capture the on/off nature of the stutter, but it would be somewhat random, somewhat variable.

[C Bradley]

I'd counsel against going the LDR route.  While their slowness to respond is ideally suited to other functions, they are too slow to be "choppy".

I'd like to see a plot of what an LDR/LED setup looks like when being driven by a square wave. What would the rise/fall time on a LDR be? Is the waveform a sine wave, triangle wave, or like a clipped sine wave?

Another question: can I drive a 2N2222's base directly from a 555 timer, or do I need a current limiting resistor? I've also got a photo transistor, which might be cool to use in a circuit with an LED, maybe a wah type of thing? Oh! do you think I could do a scratchless wah mod with the photo transistor?!?

Chris B