All the gain in final 5% of knob issue, video added

[Reg18]

I have been experimenting with 2 Lovepedal Tchulas in one box with a "stupidly simple tone control" and volume control on one and just a gain knob on the other.
First issue is the gain on the second one seems to go from nothing untill the last 5% of the turn and then it ramps up significantly.
Second issue is that the volume is significantly lower on the one with the tone control and volume which I've read is normal, but is there a simple component swap I could do to get more volume out of it?

[Rob Strand]

A similar problem come up the other day.  It's likely the load you are driving is low impedance.   The low impedance make the volume go weird will also affect the tone control frequencies.

[Reg18]

I have added a tone stack bypass, essentially putting a switch between the schematic output (which is after the diodes) and the added tone stack.
I've obviously done something wrong though as the volume and tone controls still seem
To function somewhat when in either mode. I notice the volume control alters the tone depending on where it is set, at full volume it has less top end, and at half volume it is brightest. I can an extremely wide range of tones out of this though, it's a lot of fun to play with, 2 different sounding boosts on there own, a good overdrive and an imploding fuzz sound with full gain on both.
Just need to refine the tapers/pot values etc

[anotherjim]

When I look at this kind of network, I try to see it in a purely resistive way. That means, at some frequency, the capacitors have a negligible impact on the resistive components so are ignored.
In a low pass filter like this, the caps will be open circuits whilst if it were high pass, the caps would be short circuits.
Here, you're left with a variable voltage divider operated by the 10k Tone pot.
Then there is the 1;10 rule. If a series resistance is followed by a parallel resistance that is about x10 greater, it has negligible volume reduction of signal at the junction (the top of the Volume pot). Obviously, there is a small reduction, but human hearing is not calibrated laboratory equipment!
So we see depending on the Tone pot, the series resistance goes from 4k7 to 14k7 and the 100k volume pot is close enough to x10 greater (you can often get away with x5 in guitar land). Therefore, I'd say these component values were well chosen.

Now the 100k Volume pot. Whatever it is feeding will ideally be a load of x10 = 1M. It can't be much lower than x5 = 500k. Standard guitar inputs are 1M loads, so 100k was well chosen for the output control.

Then, you have a log taper pot. If the load on the wiper contact of any pot is less than x10 the total, it will impart a non-linear action on the pots sweep and can in fact turn a linear taper in to a log taper. This is what Mr Strand is telling you above. It already is a log taper and the following load is likely making it even less linear. Furthermore, at maximum volume, the following load is now affecting the Tone control resistors.

You might "fix" the Volume pot "feel" by changing to a linear volume pot, but that will still change the Tone circuit behaviour.
Ideally, you would add a buffer circuit after the volume pot wiper to ensure it sees a 1M load.

[amptramp]

One design ideal is to have an input buffer and an output buffer with all the controls between them.  That way, there is no load or input sensitivity and the controls behave exactly the same way no matter what is connected to them.  It may add a few parts but it can stop incompatibilities between pedals.

[Reg18]

Yes I think I'd like to try this, does anyone have a link to a simple transistor buffer schematic they think would be suitable for this?
I see you can have op amp, jfet buffers as well but I have plenty of silicon transistors around but no jfets so would ideally like to use o e of those.

Can anyone shed light on why the volume knob would alter the tone? At full volume it is slightly duller than when you dial it back to half.

[Reg18]

Video added of the pedal, the boost on the right is the one that has all the gain in the last 5%. This bit is shown at the end.
https://youtu.be/jdFV5pnRIb4

[antonis]

does anyone have a link to a simple transistor buffer schematic they think would be suitable for this?
I see you can have op amp, jfet buffers as well but I have plenty of silicon transistors around but no jfets so would ideally like to use o e of those.

Something like this, perhaps..??


Although I presume you'll need a bootstrapped Emitter follower..

[anotherjim]

Copy the buffers from a Tube Screamer?

You will need the 4V5 voltage divider from R32, R33 and C16. If the buffers are only connecting to circuits inside your project, you don't need the series resistors R1 & R8. Q1 & Q3 can be any decent audio NPN BJT like 2N5088/2N5089.

Another one people copy a lot is the output stage of a Big Muff.

Simple BJT buffers don't easily manage an ideally high input impedance. Bigger resistors than shown cause more hiss. Around 500k is good going but you can get away with that, usually!

[antonis]

IMHO, for a discrete Vbias design, R32 value should be around 18k-22k due to voltage drop across R2/R13 bias resistors and Q1/Q3 V_BE..
As it is, Emitter sits on about 2.75V which might be OK for signals of lower than this amplitude but it shouldn't be a proper design for an "unaffiliated" (stand-alone) Emitter follower bias configurstion..

P.S.
In the expence of an extra cap, R2/R13 could be bootstrapped and their values could be lowered by a factor of 10..

[POTL]

Hello. In an amplification circuit, the first thing to notice is the taper of the potentiometer. One of the most popular Fuzz Face mods is replacing the potentiometer with a logarithmic one, the same type of potentiometer used by ZVEX in BOR. For volume control, try a larger value, for example 500kΩ or 1MΩ I can run a simulation in the evening and see how this affects the circuit. And yet, diodes take some of the volume.

[anotherjim]

IMHO, for a discrete Vbias design, R32 value should be around 18k-22k due to voltage drop across R2/R13 bias resistors and Q1/Q3 V_BE..
As it is, Emitter sits on about 2.75V which might be OK for signals of lower than this amplitude but it shouldn't be a proper design for an "unaffiliated" (stand-alone) Emitter follower bias configurstion..

P.S.
In the expence of an extra cap, R2/R13 could be bootstrapped and their values could be lowered by a factor of 10..

I would stick a silicon diode in series with the lower resistor. That saves working out the different resistors and keeps track with varying supply volts.
Also, I don't think you can improve headroom by any noticeable amount by worrying about a junction drop difference.
This side of the power amplifier, a little asymmetric distortion often does no harm - and the signal isn't symmetrical in the first place.

The bootstrap BJT buffer might be good and easy, but please post a scheme. I think Reg18 is new to some of this information.

[antonis]

What Jim said: 



For a second BJT, wired as Darlington configuration, a second 1N4148 in series with the existing one should be needed..
(unless the second BJT is a p-n-p compound type - Sziklai pair..)

Voltage divider resistor values shouldn't be lower than 47k 'cause their parallel combination (23k5) is set in parallel with 4k7 Emitter resistor, via 10-22μF bootstrap cap..
(the higher the h_FE the higher the voltage divider resistors value hence the lighter the Emitter loading..)