has encouraged me to start a new thread discussing it.
I wish of course to start the discussion on a good note,
with a balance of fair, generous allowance to all interested parties,
and an avoidance of any personal attacks on all sides.
To me this should be a sincere information exchange and investigation,
as well as a research project that can advance everyone's understanding of tube circuits,
for the obvious purpose of improving and selecting designs, methods, and strategies.
So before I begin diving in and shredding various design proposals,
I will happily acknowledge some important points:
(1) Many 'Aikido' designs and circuits certainly function reasonably wellPerhaps if we begin with this fair overview of the situation,
and perhaps better than less carefully crafted plans. My purpose here is not
to gainsay or contradict positive experiences of others who have built some
of these circuits, but to advance knowledge and improve all design strategems.
(2) John Broskie has certainly done all DIYers and students of tube
circuits many a service in his public offerings on the internet,
for over a decade. Anyone can benefit from reading his many online
articles and comments. My purpose here is not to in any way attack
the integrity, honesty, or talent of this generous contributor to tube lore.
So it is only in the light of these two statements of fact above,
that anything that follows should be interpreted.
level headed analysis will prevail in all the exchanges and discussions that follow.
Nothing would be more enjoyable to me than to have Mr. Broskie himself join us in the discussion.
Here was my original thesis, minus the Ace Ventura reference:
Originally Posted by Nazaroo
The comedy is in this:
It effectively cancels power supply noise when there is no signal.
When there is an actual signal, it no longer cancels power supply noise!
Its the perfect comedy, because the louder your music, the more it drowns out the power supply noise.
Effectively, you have what is called 'masking',
and it works very well.
and I believe integral element in the 'Aikido' design,
which John Broskie has applied across several different preamp/amp circuits,
made for a variety of purposes.
This is a technique whereby John cancels out
power-supply (PS) hum and noise entering a prior stage,
by applying an appropriately scaled copy of the noise signal
(inverted) to an amplifying device in the next stage,
mixing it with the original signal (containing the noise),
and thereby cancelling it out.
The (forward) feedback for this correctional system is near instantaneous,
and providing the DC current/voltage of the sample-point stay as described, (i.e., the prior stage output port),
and the noise/hum signals remain balanced (initial + copy),
cancellation occurs and noise-hum near-disappears.
That is the basic theory.
For it to hold, several things are required, and some are not.
(1) What is NOT required, is that the noise/hum be constant in amplitude or predicable in content.
(2) What IS required, is that the same noise/hum signal be present in both sources
(original and copy) on an instantaneous basis.
(3) What IS required, is that the same Amplitude (adjusted for amplification factors in each stage)
be present, for signals to cancel.
(4) What IS required, is that the same Phase for frequencies of interest is maintained
through the system, so that exact copies of the random noise-wave are reproduced, added and cancelled.
(5) What IS required, is that both (all) stages have the same 'Amplification Curve'
or Compression effects, so that signals remain balanced at all volumes.
That is the opener.
It remains for us to take a few example circuits,
and also trace a bit of the history of John Broskie's ideas,
as this technique has evolved.