Controlling the amplitude of the output signal from a pedal?

[YurkshireLad]

I'm slowly building a very basic distortion pedal from the ground up, trying to learn some basic electronics as I go. I'm a programmer by trade, so my electronics skill level compared to learning how to bake, is like knowing where the kitchen is.

I have a basic gain stage as shown in the diagram, with a gain of 11 and a power supply of +9V DC. I verified the input signal from my guitar and the output signal from the op amp (TL082) using my oscilloscope. The input from my guitar is about 250mV p2p and the output signal is boosted as expected. I didn't configure enough gain to distort the signal but I will do that later (and add transistors to clip the output) once I'm happy with a basic boost stage.

I plugged the output into my small Peavey practice amp's clean channel and set the volume to around 2. When I played my guitar, I could hear the signal through the Peavey, but the signal broke up when the guitar volume knob was set turned up. I guess it was overloading the amp and it sounded terrible, but not distorted. Perhaps the bass frequencies are too strong. I will learn how to add tone control in a bit.

Do I need a resistor on the output, after the capacitor to lower the amplitude? Is there a max voltage the output should be, say less than 1V? I only have a 10k potentiometer (need to order some) so I could put that at the output to control the volume. Is this the correct way to go or is there something fundamental I'm missing? I'm not even sure I'm asking the right questions here.

I've looked at some schematics for existing pedals and they have a volume pot at the end of the tone stage or output buffer. Any help is appreciated. Thanks

[iainpunk]

such a 10k pot is perfect for an output volume control. just hang one outer lug on the output capacitor, the other outer lug on ground and the middle lug (wiper) goes to the output jack.

there are pedals called 'boost pedals' that provide an amp with up to 9v of p2p signal.
most solid state amps react badly to being boosted in such a manner as you are probably overdriving the internal chips, which is obviously different than overdriving tubes (although there are plenty chips which sound good when overdriving and ''sounds good'' is subjective).

but the signal broke up when the guitar volume knob was set turned up ... and it sounded terrible, but not distorted

a bit of a contradictory statement here, does it sound blown out or like its starving?

welcome to the forum,
cheers

[idy]

Welcome to the forum!

When this is in context, followed by other stages, requirements may change. Generally:
Yes, some small (maybe 1k?) resistor in series with the output is good practice just to avoid a possible short of the output. That will limit current. To lower voltage you need two resistors or a pot. (look up voltage divider.)
Yes a volume control will give you a voltage divider and allow you to keep your amp from being overwhelmed.
Tone controls are usually treble cut (low pass). The basic idea is bass cut early, before you drive the boost stage into cut-out, and treble cut after the distortion stage to smooth things out.
So first try changing the input cap from 1uf to something like .022uf, like a tube screamer. Try to find a value that thins things out right for you.
10uf output cap is also overkill. Maybe 1uf there.
Best level for your amp depends... Tube amps are generally happy with pretty big signals (some guys like pedals that run on higher voltage, up to 27v like the Klon.) Transistor amps are not famous for loving high voltage boosters.

[antonis]

Hi & Welcome, also..

23k (56k//56k) input impedance isn't "ideal" for such a configuration,,

[YurkshireLad]

such a 10k pot is perfect for an output volume control. just hang one outer lug on the output capacitor, the other outer lug on ground and the middle lug (wiper) goes to the output jack.

there are pedals called 'boost pedals' that provide an amp with up to 9v of p2p signal.
most solid state amps react badly to being boosted in such a manner as you are probably overdriving the internal chips, which is obviously different than overdriving tubes (although there are plenty chips which sound good when overdriving and ''sounds good'' is subjective).

but the signal broke up when the guitar volume knob was set turned up ... and it sounded terrible, but not distorted

a bit of a contradictory statement here, does it sound blown out or like its starving?

welcome to the forum,
cheers

Hmm, I'm not sure of the auditory difference, but perhaps blown out? Mushy? 

[YurkshireLad]

Hi & Welcome, also..

23k (56k//56k) input impedance isn't "ideal" for such a configuration,,

What do you recommend? I will look at sample circuits again to see what they use. Thanks.

[idy]

Actually, with your relatively low input impedance (those 56k bias Rs) you may not need as small as 22n. But try it.
The lower the input impedance, the bigger that cap needs to be. See fuzz face (2.2uf), rangemaster (5n and cuts lots of bass) vs high impedance MOSFET boost like SHO (1n and full range).

[antonis]

What do you recommend?

Either 10 to 20 times higher value or a 1M extra resistor going from 56k/56k junction to non-inverting input..
(for the later, voltage divider junction should be set out of signal path, of course..) 

[GibsonGM]

I'd recc. trying the suggestions above...smaller input cap.  We find these "AC AMPLIFIER SCHEMATIC" things online, and have to tinker with them. The easy way is to steal  use work that has already been done for us.

The 2 56k resistors are providing the opamp with bias - so the AC signal can work in the DC-operated device...to increase the input impedance you could use a 'bias network' with say, 2  10k resistors and a bypass cap to get the 4.5v you need for bias, and send it to the opamp thru a  220k to 1Meg resistor, thus improving the input impedance.

Sample Circuit:  An example from the MXR Distortion Plus...schematic HERE https://scfxguide.wordpress.com/tag/distortion/   You would want R2 and R3 to be 10 to 22k or so, though.  The 1Meg feeding the opamp can be lower, that is common with this 'starved' design...   Just an example of what the 'norm' is with biasing an opamp (if there is one).  good work so far!   There are many ways to skin this cat. 

[YurkshireLad]

What do you recommend?

Either 10 to 20 times higher value or a 1M extra resistor going from 56k/56k junction to non-inverting input..
(for the later, voltage divider junction should be set out of signal path, of course..) 

Oh, so connect the 4.5V rail to the input signal path, instead of running the input signal through a voltage divider? Like in the Boss DS-1?

http://www.electrosmash.com/images/tech/ds1/boss-ds1-distortion-schematic-parts.jpg

[GibsonGM]

Yes, R10!   

[antonis]

MS Draw sukcs..

[YurkshireLad]

I managed to put this together in LTSpice - I may have screwed up some of the values, but it's a start. It's actually fun simulating the guitar signal and then probing around the circuit. 

[GibsonGM]

Nice!  Just a couple of fixes...Spice doesn't care what you do, it's stupid and will give happy returns even if we mess up, so...

There are rules for bias supplies (I don't have at hand, I have had a few pints! 10 times, blardy blar)...100k is too high a value for the divider.  Try 22k or 10k, as a rule.  10k (well, 20k total)  is only a fraction of a mA anyway, so no worry! 

You obtained 4.5v bias, great!  Now...to feed it to the opamp, look at Distortion Plus or the FULL DS-1 schematic...where you have 4.5V input to the opamp, there is a resistor.   It is there cuz direct connection would cause AC to ground thru the CAP in the divider!  AC and DC are separate and different issues.  The resistor keeps AC from 'flowing backwards' up the pipe to the divider.  You will build up to understanding as you do this.    Add a 1M, or 470k, etc, resistor between the tap of the divider and the opamp input.  It is a big player in the input impedance mentioned earlier, so between those 2 values will put us in a good place.  Pick something, lol.  See what you get, and we'll visit this more.

Good first steps!

[GibsonGM]

Some light reading that helped me out...Pictures of Lily for the opamp enthusiast!  Can't do 'the thing' we want to do until we know how to use the device(s) that we need to do them with, and all that.  FYI, we're mostly "split supply", so check that stuff out early...


https://mil.ufl.edu/4924/docs/TI_SingleSupply_OpAmp.pdf

[Vivek]

1m is 1 milliohm in SPICE

1Megs is 1 Mega ohm

Suppose you have gain 11

And input 1vp

You expect 11vp output

Which is not possible with 9v supply

But you won't see that in spice if you use model of ideal or generic Opamp which has no connections for Vcc

So please use model for TL072 etc where you need to connect Vcc to the Opamp model.

[YurkshireLad]

1m is 1 milliohm in SPICE

1Megs is 1 Mega ohm

Suppose you have gain 11

And input 1vp

You expect 11vp output

Which is not possible with 9v supply

But you won't see that in spice if you use model of ideal or generic Opamp which has no connections for Vcc

So please use model for TL072 etc where you need to connect Vcc to the Opamp model.

Thanks. I didn't know how to specify mega ohms but now I know it's "1 meg".

Hmmm. I thought I looked for a TL072 but couldn't find it. I'll look again.

[Vivek]

LTSPICE model for TL072


http://ltwiki.org/index.php?title=File:TL072.sub

[YurkshireLad]

LTSPICE model for TL072


http://ltwiki.org/index.php?title=File:TL072.sub

Thanks I'll, install it later.

[antonis]

Just take into account what Sir Mike said..

As it is, input impedance is about 50k (100k//100k) and in case of a V_Bias cap, it simply is zero..
(OK.. not absolute zero but very close to nothingness..)

P.S.1
You tend to place input electro caps with wrong polarity..

P.S.2
You set circuit ground one diode voltage drop above PS ground..