Box of rock clone, volume problem

[ricci]

Now I'm posting the images but the values are different, in the second there is a square wave and in the third the beveled square

[ricci]

Images here

[Helmholtz]

Your input signal is way too large to tell gain. Average guitar signal is significantly lower.
With a square wave output the term gain makes no sense.
Lower input signal until the signal at the third drain resembles a sine.
If necessary use a voltage divider or pot before the input.

[ricci]

New images. I'm not very familiar with the scope, I had to fiddle with it a bit before seeing an input signal in the mV range.

Edit:

All controls are at maximum, including gain

[Helmholtz]

I assume the first pic is the input signal.
Being around 400mVpp it's still too large for an average PU signal.
Let's forget about determining exact gain.

Using this input signal you should be able to see differences between the pedals.

[ricci]

Yes, first pic is input. Now, images d1, d2 and d3 before my clone and after the best performing one.

[Helmholtz]

Confused.
Please clearly identify which pic was scoped where.
Why are the first 3 pics different from the ones posted before?
Are you using a X10 probe? Makes no sense here, just increases noise and lowers sensitivity.

[ricci]

They are slightly different because I had switched off for a moment
Yes 10X, now 1X.

Order of pics

Input,
d1,d2,d3,
d1_perf, d2_perf, d3_perf

Good evening to you and good night to me

[ricci]

You checked tolerance on resistors and caps, but how about the pots? If they are not precision type they can have quite wide variations.

A few days ago, thinking about some defect in the pots, I tried to remove them, shorting the central lug with lug 3 to have maximum control but nothing changed and the gap was still there.

[x-pro]

... all Alpha brands and identical also in values, of course.

and the characteristic of resistors?

[ricci]

and the characteristic of resistors?

what kind of features? They are 2% wire resistors.

[x-pro]

we're talking about variable resistors

[ricci]

The only variable resistors in the circuit are the pots and I imagine you are referring to these, the pots are the same as the other pedal and the original pedal,
Volume 100k logs
100k Lin Tone
Gain Antilog 5k.

They all work well, in fact by removing them and shorting lugs 2-3 the gap problem remains.

I also thought about some cold soldering, but after a reflow nothing changed.

And it's not even the PCB's fault because I've made this blessed pedal 3 times on 3 different PCBs, even Point to Point but it always works with the same problem.

[x-pro]

ricci, welcome: https://guitar-gear.ru/forum/topic/5...k/page__st__50

[Helmholtz]

No signs of circuit issues.
Signal levels in scope pics are not significantly different. Signal shapes are a little different, though.
Not sure if this can explain the perceived loudness difference assuming that the scope pictures were taken with the same pedal settings as used for sound comparison.

But - as said earlier - distorted and clipped waveshapes don't allow to compare real guitar level signal gain.

Some theory:

The drain to gate voltage NFB results in:
1) Low input impedance, maybe in the 10k range.
2) The gain of the input stage depending on connected source impedance.
Lower connected impedance results in less NFB and thus higher gain.
So a low impedance signal generator will see considerably more gain than a guitar PU. (Might be interesting to test with a 100k resistor in series with the sig.gen.).

Both effects will change the frequency response of a guitar PU which has low impedance at low frequencies and high impedance at treble.

This behaviour is totally different from a tube input stage.
In fact, when I once tried a Z.Vex BOR, it changed the sound character of my guitars considerably.
After a few days of intense testing I returned the pedal.