Simple op amp distortion circuit.

[Dragonfly]

This will be "old hat" for the regulars here, but it's really aimed for the beginning builder. Sometimes op amps can intimidate new builders, so getting a really simple circuit or 2 under your belt can give you the confidence to go off and build more complex circuits.

Anyway, this is a very easy to build, good sounding distortion. It has a low part count, and all of the parts should be easy to find. Its similar to things like the Dan Armstrong Blue Clipper, and the MXR Distortion +, but i tweaked it for a bit fatter sound and more gain on tap. It cleans up decently with your guitars volume knob as well. It will still clip if you remove the diodes, but I like them in the circuit.

Also, I've included a sound sample, though I dont think the sound sample even comes close to doing it justice. Its a bit less buzzy, and the bottom end is a bit more pleasing, IMO.

Without further adieu....

CLICK HERE FOR SOUND SAMPLE....NOT THE BEST DEMONSTRATION OF THE SOUND, BUT IT GIVES THE IDEA

[GibsonGM]

Hey Dragonfly,
Nice job on that one!  Makes me want to Simple Clipify my Dist +; sometimes simple is better!
Like the clip, too. 
) 

[Barcode80]

Andrew, I've made a fairly compact PCB for this if you don't mind me posting it. Might be good for those wanting to start etching pcbs (as all my layouts are, ost are very simple circuits! )

[Dragonfly]

Andrew, I've made a fairly compact PCB for this if you don't mind me posting it. Might be good for those wanting to start etching pcbs (as all my layouts are, ost are very simple circuits! )

Feel free to post it

[Barcode80]

well, you know, i like to ask first



the pnp transfer is in my gallery, page 3, print at about 200 DPI.
http://aronnelson.com/gallery/BarcodeVeroLayouts
btw everyone, check out my PCB layout gallery! I'm up to 3 pages now LEAVE COMMENTS IF YOU BUILD!! 

[Mark Hammer]

A good simple starter circuit.  Historians will note that both the Dan Armstrong Blue Clipper and the E-H Muff Fuzz were similar sorts of distortions in that they had a preset gain/drive/fuzz and the user could only vary output level.  So, this carries on a noble tradition of simplicity in design.

Four points of note:

1) The circuit is inverting.  That means that if you try to mix it with a noninverted signal, there will be cancellations.

2) It uses an interesting approach to tone shaping by having a fairly high rolloff on the input (starts rolling off below 3.4khz by my calculations), and a fairly low high-end rolloff on the output (starts trimming treble above around 480hz by my calculations).  The 330pf feedback cap is why you don't see a .001uf cap to ground alongside the diodes, like you would in a Dist+.  You can get more bass if the .047uf input cap is increased a bit (e.g., 0.1uf/100nf or even .22uf/220nf) and varying amounts of sizzle by either decreasing the 330pf cap (for more treble) down to maybe 220pf or so, or by adding a cap in parallel with the diodes (e.g., .001-.0022).

3) The circuit has a gain of x1000.  For those who want to have a sense of what different gains get you in terms of clipping, this is what x1000 sounds like.  This also means that some of the clipping produced is coming from the 741, and not just the diodes.  It also means you could stick a pair of LEDs in parallel with the 1M/330pf combo, clip the bejeezus out of the signal and still have enough juice to produce additional clipping when the output hits those 1N914s.

4) It should probably include some sort of power smoothing cap of some sort, like a 47-100uf cap from V+ to ground. (I commit the same sin, leaving it off far too often).

[Gus]

The circuit only has a gain of 1000 when driven by a 0 ohm z voltage source or after a buffer or buffered effect with low output Z

The circuit is clever it is kind of like a FF the input is a summing node

The 1 K ADDS to the source Z the guitar or bass mess in front of the circuit

this also causes a EQ curve because of the inductance and capactance and resistance mess the passive guitar or bass is adding.

The guitar or bass is part of the feedback calculations with inverting circuits like this.

  So you don't know the time constant unless you know the guitar or bass mess inside.  you switch pickup it changes etc.....  add up the Rs and L and C then the coupling caps or better just build it and try it out

I have posted about stuff like this FF threads (imperfect  summing node this is why people select the Hfe of the first transistor and building an inverting opamp fuzz))and the gruntbox having a 10K fixed input possably acting different with a active bass or after an effect.

I sometimes put active/passive switches on FFs the active adds a series pot like the Fuller mod the passive is no series resistor to the input node

[Mark Hammer]

The circuit only has a gain of 1000 when driven by a 0 ohm z voltage source or after a buffer or buffered effect with low output Z

The circuit is clever it is kind of like a FF the input is a summing node

The 1 K ADDS to the source Z the guitar or bass mess in front of the circuit

this also causes a EQ curve because of the inductance and capactance and resistance mess the passive guitar or bass is adding.

The guitar or bass is part of the feedback calculations with inverting circuits like this.

  So you don't know the time constant unless you know the guitar or bass mess inside.  you switch pickup it changes etc.....  add up the Rs and L and C then the coupling caps or better just build it and try it out

I have posted about stuff like this FF threads (imperfect  summing node this is why people select the Hfe of the first transistor and building an inverting opamp fuzz))and the gruntbox having a 10K fixed input possably acting different with a active bass or after an effect.

I sometimes put active/passive switches on FFs the active adds a series pot like the Fuller mod the passive is no series resistor to the input node

See, now this I did not know.  I grew up thinking that gain = Rfeedback / Rinput and that was it.  From what you've indicated, that is true for stages within a circuit but not necessarily front ends.  So you're telling me the gain is actually less than x1000?
So how does one stick an inverting stage at the front and have any sense of what the frequency response or gain will be in advance?  I'm not disagreeing with you, I'm just asking for a way to begin making the necessarily calculations.

[rankine]

Looks like a great little circuit to play around with. Thanks for sharing it with us.
To the breadboard! 

[johngreene]

See, now this I did not know.  I grew up thinking that gain = Rfeedback / Rinput and that was it.  From what you've indicated, that is true for stages within a circuit but not necessarily front ends.  So you're telling me the gain is actually less than x1000?
So how does one stick an inverting stage at the front and have any sense of what the frequency response or gain will be in advance?  I'm not disagreeing with you, I'm just asking for a way to begin making the necessarily calculations.

gain = Rfeedback/Rinput still holds true. The source output impedance has to be included in Rinput. If you want to have predictable results with various input source impedances, then you need to buffer it.

--john

[Gus]

You need to think of the inverting stage - input as a current summing node it is closer to perfect with an opamp than a FF

now the gain of 1000 is what you posted IF the source feeding the input R (1K here) is 0 ohms in other words a perfect voltage source.  So if the source is something like a guitar one would model it as a voltage source with the added inductance resistance capactance in series with the 1K to ground

in case this gain would be as it varies with freq

1meg and the cap 330pf divided by (1K + all the stuff in the guitar)

then because of the inductance and capactance there will be a freq depended gain or as I posted an EQ that will change with the vol and tone pot of the passive guitar or bass.

you need to add up the guitar and 1K to figure out the highpass filter freq with the .047uf it can be a mess so trial and error might be the best way and it will change with guitars and basses that have different passive circuits.  Change the pickup switch tone and vol and it changes.

That EQ is also a big part of getting the FF setup as one wants it

Now when one makes the resistors larger the source impedance has less effect but you get more noise sometimes and other things happen.

this is part of the the FF after the wha with a lower output R problem

In the FF case the input R( 1k in this circuit) is the guitar or bass

If people don't understand my post ask

Maybe R.G. or someoneelse can word it a different way that others understand

It is all trade offs

I wondered why the gruntbox thread died, maybe people did not understand may post.  Lower value input Rs to an inverting opamp

Anymore questions ask.

[Dragonfly]

See...its things like this that i LOVE about this forum...I post a simple beginners style circuit that i came up with by messing around on the breadboard, and i end up learning some new things. 

==================================================================================

Mark...having the input cap above .068 gets a bit bassy in this configuration, though i imagine that decreasing the output cap and the feedback resistor a bit would tend to help combat that. As for the 33pf cap, i tried everything from a 50pf to a 330pf in the loop, and a .001 in parallel with the diodes, and i thought that anywhere from a 220pf-330pf sounded best. Again, with this particular configuration. Changing input / output caps would probably change my preference. And you are correct...a 47uf-100uf cap to ground from the power supply is always a good idea.

Gus...I don't know how you do it, but whenever you post something, i learn something new. Thanks. Sincerely. RG is the same way for me...you guys have a way of explaining things that still makes sense for a "non-techie" like me. I had never thought about how a small value resistor would allow the instrument to have a greater effect on the input impedance. Definitely something to mess with. IIRC, didn't Joe Gagan do something similar utilizing a pot on the input of some of his FF designs ? (variable impedance?)

John...I thought about adding input / output buffers (JFET), but in the interest of keeping things simple and "unintimidating" for the beginner, i thought it best to leave them off. I thing the results would be a bit better WITH the buffer however.

Rankine...Put on on the breadboard and play around. One of the good things about this circuit is that its fairly easy to hear the changes you make. Tweaking it for your personal taste is one of the great things about DIY.

==================================================================================


A nice mod for this is to use a 1M pot in place of the 1M resistor in the loop...or even better, a 100k resistor in series with a 1M pot. That gives you a simple gain control.

[Gus]

Something different for a change not an 808 or ff etc and the thread dies?   

This circuit and ones like the gruntbox are good ones to build on and learn from.

The FT 69 with the input pot should make sense to more people now

Now what is this web thing about being 1/2 supply voltage.  For fun make the "bottom" 47K   1K 10K 22K and/orincrease the top 47K  I have done stuff like this years ago

Offset the opamap "up" or "down" via the ref voltage

try a 5334 or a TL071 or............

remove the clipping diodes for just opamp distortion

try different input resistors for the 1K 100,220,330,470,680,820,1.2K etc

This is a cool little circuit to learn from.  Simple low parts count BUT has things that are good to understand about opamps used like this and even the FF

[Dragonfly]

Something different for a change not an 808 or ff etc and the thread dies?   

This circuit and ones like the gruntbox are good ones to build on and learn from.

The FT 69 with the input pot should make sense to more people now

Now what is this web thing about being 1/2 supply voltage.  For fun make the "bottom" 47K   1K 10K 22K and/orincrease the top 47K  I have done stuff like this years ago

Offset the opamap "up" or "down" via the ref voltage

try a 5334 or a TL071 or............

remove the clipping diodes for just opamp distortion

try different input resistors for the 1K 100,220,330,470,680,820,1.2K etc

This is a cool little circuit to learn from.  Simple low parts count BUT has things that are good to understand about opamps used like this and even the FF

Didnt the Blue Clipper or something like it bias the opamp using mismatched resistors ?

Your post definitely helped my understand the input impedance issue a bit more, and how you can help tailor the sound to a particular instrument

The opamp distortion definitely adds to the "rawness" of this circuits sound

I need to go have a look at the gruntbox...sounds interesting.

and yeah, its a shame that this thread sinks to the bottom so quickly.  maybe it will get rediscovered after a while.

.

[markm]

Something different for a change not an 808 or ff etc and the thread dies?   

Welcome to my world fellas. 

[Gus]

Something else to add

20 db X10
40 db X100
60 db X1000

Now look at the open loop gain curves of Opamps(check the NS site etc)

A 741 little over 100 db openloop gain up to about 10Hz and about 60 db openloop gain at about 1KHz and about 40 db at 10KHz

So you can only have 60 db of closed loop gain to less than 1KHz

Other opamps move the low frequency corner up a 5534  about 400 to 800Hz(comp cap) and still has >60db open loop gain at 10Khz

The way opamps are sometimes used in effects EH muff fuzz MXR, Dist +, this circuit..... the open loop curve might help understanding why some people like older ICs like the 741 sometimes.

  The open loop curve  is a low pass filter effect (integrator) read the "class a gain stage section" at http://en.wikipedia.org/wiki/Operational_amplifier

[Dragonfly]

Something else to add

20 db X10
40 db X100
60 db X1000

Now look at the open loop gain curves of Opamps(check the NS site etc)

A 741 little over 100 db openloop gain up to about 10Hz and about 60 db openloop gain at about 1KHz and about 40 db at 10KHz

So you can only have 60 db of closed loop gain to less than 1KHz

Other opamps move the low frequency corner up a 5534  about 400 to 800Hz(comp cap) and still has >60db open loop gain at 10Khz

The way opamps are sometimes used in effects EH muff fuzz MXR, Dist +, this circuit..... the open loop curve might help understanding why some people like older ICs like the 741 sometimes.

  The open loop curve  is a low pass filter effect (integrator) read the "class a gain stage section" at http://en.wikipedia.org/wiki/Operational_amplifier

GREAT info Gus...

Thanks again...

AC

[Mark Hammer]

Conceivably one of the reasons why there was no feedback cap in the Dist+ in parallel with the 1M feedback resistor; the chip itself acted as a bandwidth constraining element once the gain started to go up.

[theonlyrobkeyser]

I've been trying to get a better understanding of these types of circuits and this is a great thread.

I have a question about using the inverted input of the IC.  Why would you use the inverting vs. the non-inverting?  The Dist+ uses the non-inverting, but this circuit is inverting.  If that inversion potentially could create problems down the road with cancelations, why would you do it?

Thanks - Rob

[WGTP]

MUFF FUZZ op amp version + Gus' Mods