reverse Beta 2n2222 Harmonic Instant Coffee (Percolator modded)

Started by iainpunk, January 28, 2021, 09:12:11 PM

Previous topic - Next topic

iainpunk

so i was playing around with a Bazz Fuss and harmonic percolator ''''hybrid'''' design. for Q1 the PNP i just used a BC557b since it does its job well enough, and different transistors seem to have little impact on the sound. Q2, however, has way more impact on the final sound, some Si transistors give a riff-raff, agressive, slightly octave-ish sound, others sound really fat, over compressed and huge, until i tried the 2n2222 in reverse Beta (C and E swapped, this gives roughly 1/5th of the Hfe) it sounded really gnalry, octave and gated at the same time, the perfect nasty fuzz sound, it sounded beautifully broken. but the diode i plugged in was not fully seated and i knocked it loose, and out came the biggest, thick sustaining, full and smooth fuzz sound, the opposite end of the tonal spectrum, when it comes to fuzz.

what this means it that the reverse Beta 2n2222 without any bias circuitry biases it self, and sounds really thick?! does that come through leakage, or something else? other transistor's don't seem to work without the biasing elements.

thanks in advance for your insight,
cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

iainpunk

#1
EDIT: schematic
i noticed a mistake in the schematic, this one also has the extra diode so it works with every transistor:


i call it the harmonic instant coffee....
percolator coffee = good, smooth, full
harmonic percolator = good, smooth, full
instant coffe = bad, sharp, thin
harmonic instant coffee = bad, sharp, thin

the PNP is a BC557, the NPN is a 2N2222A and the diode is a BYX55 BYX10.
there was another LED on the NPN the same way a Bazz Fuss has its diode, having that diode (LED) or not makes a huge difference in sound.

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Keppy

"Electrons go where I tell them to go." - wavley

nocentelli

Quote from: iainpunk on January 28, 2021, 09:12:11 PM
the diode is a BYX55 BYX10

This is the lower diode? How critical is the part listed?
Quote from: kayceesqueeze on the back and never open it up again

iainpunk

#4
Quote from: nocentelli on January 30, 2021, 04:39:49 AM
Quote from: iainpunk on January 28, 2021, 09:12:11 PM
the diode is a BYX55 BYX10

This is the lower diode? How critical is the part listed?
both diodes are not critical at all, a good replacement is the 1n400x, but 4148 or Ge diodes work too, even white LED's work, but they all have slightly different sounds. i just liked the BYX10's sound the best for the bottom diode, i use a super bright white LED for the top one, but its only a small difference. the top diode's larges difference is the output volume.

when using a reverse Beta 2N2222A the top diode can be left out in my experience, but your mileage may vary.

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Digital Larry

The Harmonic Percolator is one of those "mystery" circuits for me.  It does not seem to be based on any standard topology used anywhere else for any other reason.  It also sounds pretty rad!  I may just build one one of these days.  This reverse transistor idea is fascinating too.  So many of the designs seem to prefer lower gain transistors which are harder to find.  The only problem is that my transistor tester figures out which lead is which automatically so I'm not sure how I'd confirm what the gain diff of backwards installation would be.
Digital Larry
Want to quickly design your own effects patches for the Spin FV-1 DSP chip?
https://github.com/HolyCityAudio/SpinCAD-Designer

mac

All Ge version with octave.



QuoteSo many of the designs seem to prefer lower gain transistors which are harder to find

MPSA42 is low gain, <150, cheap and easy to find.
BDxxx and TIPxx too.

mac
mac@mac-pc:~$ sudo apt-get install ECC83 EL84

iainpunk

i measure gains with a 10nA current source on the base, and use a resistor bank to check which resistance gives the best measurable voltage (near 4.5v) , and then measure the voltage, calculate the current, and then calculate the Hfe.
the bank consists of a 100r, 330r, 1k, 3k3, 10k and 33k resistors on a rotary switch.

QuoteAll Ge version with octave.
what do you mean with this? did you build one with all Ge transistors and diodes and it does an octave up effect?, that would be realistic to expect, some transistor diode combo's create an octave effect, since its basically a deluxe Bazz Fuss witht the first gain stage inverted to PNP, and then stacked onto each other, lowering the headroom.
the Harmonic Percolator is kinda simple if you break it down in to two gain stages, and the big cap in the middle reduces interplay between the gain stages, making it possible to calculate them separately.

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

mac

Quotewhat do you mean with this?...

It's been 14 years...  :o
I do remember that I had to include a resistor between emiters to bias the thing. I used a pot and I got 8up at max settings.
A funny accident.
Big chances that it works in a standard HP.
Diodes just clip.

Lower emiter sees an inverted and smaller signal than the upper emiter.
With no resistor the big cap grounds both transistors.
If a resistor is placed the emiters begin to push and pull.

mac



mac@mac-pc:~$ sudo apt-get install ECC83 EL84

anotherjim

My brain fails to see any reverse beta in Iain's scheme. The 100uF on the upper NPN emitter holds that at some half-way DC voltage won't it? So, effectively making a split supply.  I see one PNP CE amp on the lower half AC coupled to an NPN CE on the upper half.
Is it possible the 100uF is across the middle 1k?


iainpunk

Quote from: anotherjim on January 30, 2021, 01:44:24 PM
My brain fails to see any reverse beta in Iain's scheme.
sorry, i replaced the schematic with a schematic that works with almost all transistors, not just with the reverse bias 2n2222a.
the version with the reverse bias 2n2222 just doesn't have the top diode and the C and E are reversed.

QuoteThe 100uF on the upper NPN emitter holds that at some half-way DC voltage won't it? So, effectively making a split supply.  I see one PNP CE amp on the lower half AC coupled to an NPN CE on the upper half.
Is it possible the 100uF is across the middle 1k?
no, its not across the 1k, its based on the Harmonic Percolator, that capacitor does make it a kind of split supply, just like the HP, but its also like the Deluxe Bazz Fuss.

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers