A lead boost built and tested

[Gus]

Another booster, An alternative to JFETs
Built and tested measured voltages are lower than shown in the screenshot. R5 is where you adjust the bias
Input R can be increased try 1.5meg for R3
Yes 1000pF(.001uF) or smaller input cap you often want to reduce the bass when boosting
Try different transistor numbers 2n3904 and 2n3906, 2n4401and 2n4403 NPN and PNP betas >200

100k output volume control

[Mark Hammer]

What particular aspects of this do you like, or feel add value?  Or is it simply another way of doing the same thing with different (and perhaps more available for some folks) parts?  Just curious.

[Gus]

It uses a Sziklai pair http://sound.westhost.com/articles/cmpd-vs-darl.htm worth the time to read. A web search will find more pages
It has a small amount of parts
Has a gain of about X10
It is a different than other boosts with the Sziklai pair. R.G. has a nice circuit using a Sziklai pair at GEOFEX

You often see JFET circuits with 1meg gate to ground and 1meg pulldowns/antipop resistors at the input
For guitar boost/distortions you don't need JFETs for about 1meg input resistance

A builder could adjust the input cap value to their liking I think .001uf is a good starting point

I used a 2N3904 and 2N4403 just because they were in bags on the bench.
I would try a number of Silicon NPN PNP pairs Betas >200

[Gus]

A good link

https://www.its.caltech.edu/~musiclab/feedback-paper-acrobat.pdf

pages 1 to 24 are interesting
look at the graphs on pages 12,16,18 and 22

[Gus]

Having issues getting DIYLC working correctly in Ubuntu, otherwise I would have posted a stripboard layout. It fits on a 9 x16 with the volume control soldered to the board making it easy to mount by the volume control.

This is a nice boost

[Gus]

png file from a new toy.  Note the nice distortion in the output triangle wave  Taken at 1KHz output volume at max. You can measure the gain looking at the traces and channel settings.
I find triangle waves helpful for checking circuits with wanted distortion it is also easily to see what side clips first with a triangle compared to sine
Next I will try the FFT with a 1KHz sine

[Gus]

I think this is more repeatable than an JFET boost like the EP boost
It is a very nice sounding circuit
People should also try R.G.'s circuit

FWIW when I built this lead boost on stripboard I needed to ultrasonically clean the board because of leakage currents at the base of Q1 caused by the flux. A wipe with isopropyl alcohol did not help

[lars-musik]

FWIW when I built this lead boost on stripboard I needed to ultrasonically clean the board because of leakage currents at the base of Q1 caused by the flux. A wipe with isopropyl alcohol did not help

That's interesting. Is there no harm in immersing the cicuit in an ultrasound bath? There are some cheap cleaning devices for glasses on various shopping sites and I wonder if it would be worth a try.

[Gus]

Lars

You can damage parts

google etc. "testing an ultrasonic cleaner with strips" look at what happens to test strips

There is more to using a ultrasonic cleaner than just placing parts in it

A wipe should have cleaned the board well enough for the 1meg resistor used. I did not use an epoxy glass stripboard.

[Transmogrifox]

Very nice Gus.  I made a stereo-to-mono mixer based on the Sziklai pair used like this as a super-transistor.  In the stereo-mono case I used a JFET as the input device and then a 2N5087 as the PNP part.  For mine was 10k in the emitter and about 3.5V bias, +/-5.25V rails.  Mine was optimized to <0.05% Harmonic distortion (according to SPICE model) and because it's really low noise it was probably less than 0.05% THD even though I have never measured it.

Really clean audio anyway.

I think as an intentionally nonlinear booster it looks promising.  It would be interesting to see if using a JFET instead of an NPN would make an audible difference in its clipping characteristic when it goes nonlinear.  The feedback will correct for a lot of JFET variation so the variation in JFET devices is likely to result only in clipping symmetry.

Just my musings.