AM:
In Russia, you
are known to many by your articles in
"Radio" magazine; you were also known
as a Chief Designer of the Lvov Radio-Electronics
Co. where "Amphiton" amplifiers, in
particular, had been designed by you. Many know
that you had emigrated to the U.S....
VS: For me, things had not
turned out quite favorably. While waiting for the
U.S. entry visa in Italy, I broke my ligament
playing tennis (I was a professional table tennis
player). A knee operation in America, where I
arrived shortly afterwards, was not successful;
two following years had not been too productive
for me. But, as they say, there won't be fortune
if not for misfortune: during that time I had
been working on my English and began sending
resumes to high-end audio companies. However, I
discovered that those companies needed mostly
technicians, as their own design philosophy had
already been formed and crystallized. Eventually,
I found one company, "Madison Fielding,
Inc."
Originally,
they invited me to work on simple designs but I
offered to contribute high-end designs; they
accepted. As a result, I became a partner in the
company. My first implemented designs were power
amplifiers model M1 and M2 which were exhibited
for the first time at the Summer CES in Chicago,
IL in 1993. For us, it was quite a sensational
presentation (M1s were drivingWilson
WATT/Puppy2); the tandem was labelled "the
best sound at the Show" (you probably have
this information from the materials I sent you).
However,
my partners at Madison Fielding, Inc. were not
the "high-end" material; in other
words, they did not fully grasp what this
business entailed. It was only after that Show
that I realized that the company had no intention
of starting production. Although I had
practically no resources for doing so, I left
"Madison Fielding, Inc." to establish
my own company which began its operation two
months afterwards.
That
was in October, 1993. Now, this is my sixth Show.
During the past three years I had completed and
implemented into production a line-level
preamplifier model L1 with a fully balanced
circuitry. It is a very interesting model -- I
use superlinear high voltage MOSFET transistors
in the front end and +350V vacuum tube voltage
regulator. Right now, I have a couple more
preamplifier designs (not implemented into
production yet) which are less expensive than the
L1...
AM: What technical solutions
characterize LAMM designs?
VS: I had been working for a
long time on the question of why one piece of
equipment makes good sound and the other does
not. This question is as old as the world itself.
I took the path of psychoacoustics and developed
a number of mathematical models which manifested
a more or less close approximation to reality
(fortunately, back in Russia, I had a whole
R&D department at my disposal). Those
mathematical models were described by
differential equations; specific correlation
systems were developed. Using this information I
obtained the electronic models of processes.
AM: In other words, you
created something akin to the analog computer?
VS: Correct. As a result, I
ended up with a very limited number of topologies
which can be used in audio applications. Now I
make audio equipment using those results. As a
rule, I do not go through listening tests of the
designs I create: when design is finished I make
one listening test for control purposes...
AM: Simply speaking, you know
the results beforehand?
VS: I do. Each model is
designed absolutely deliberately with
predetermined and predictable characteristics.
AM: Let's say, a Japanese
reviewer N. comes to you and says that everything
is great but he would like to have a deeper
soundstage. Do you know what you will be doing?
VS: Yes, I know what I will be
doing: making a soundstage deeper (laughing). I
set certain requirements for each of my designs
and implement them -- it is confimed by practice.
In particular, the amplifier we are looking at --
model M1.1 -- is our flagship design in a
solid-state line. It operates in pure class A
mode and its topology features no feedback. We
use high speed power MOSFET transistors.
I had
spent a lot of time figuring out how to make the
output stage work properly. It had been difficult
because I use no feedback. As a result, even no
servo system is employed and my amplifiers are
thermally absolutely stable: output DC offset is
no more than 10-15 mV in the temperature range
between 10-45°C ambient--this is a very good
result.
AM: Is there anything at the
output, for instance, inductor?
VS: No inductors. The output
impedance of an amplifier is about 0.18-0.2 Ohms,
linear from 20Hz to 25KHz; in other words, the
active component is overwhelmingly dominant in
the output impedance. That is one of the reasons
why the amplifier behaves as a magnifying glass
with many types of speakers, making clear what is
going on in a system as a whole because the
amplifier is practically neutral itself. Besides,
these amps can be used as welding machines (in
other words, the amps can function even under
short-circuit conditions). The input current is
limited only by a line fuse, the value of which
had been calculated to activate it, should the
output power exceed about 400-500 Watts. The
front end topology is quite interesting--if I
were to simplify its description I would say that
the input is a cascode differential long-tail
pair utilizing matched p-channel FET transistors
and the high voltage p-channel bipolar
transistors; the so-called Wilson current mirror
is used as a current source; 6922/ECC88 vacuum
tube is used in the second stage; the third stage
is a single-ended high-current buffer on n-p-n
bipolar transistors utilizing a current source.
At a first glance, it is quite simple; however,
as I mentioned before, these circuitries have
been developed according to the electronic models
I had already created. Here I use a quite simple,
from my point of view, biasing circuitry that
also provides an excellent thermal stability of
quiescent current of the output transistors. I do
not know who else utilizes such circuitry. I had
spent a lot of time solving this issue; of
course, there exist standard solutions for
resolving this problem, but these types of
solutions are quite costly and not always very
effective. Besides, my amplifier is expensive as
it is.
Other
features that add to the proper functioning of
amplifiers include a soft-start circuitry. For
the first 2.5 seconds voltage is not applied to
the filter capacitors since the big caps would
blow any power supply while in a transition mode
(which leads to a short circuit); therefore, a
soft-start circuitry is used; next, time relays
kick into operation.
AM: Through the resistors?
VS: Three power resistors
connected in parallel are, in turn, in series
with the mains. I use a very serious protection
circuitry, with the delay time of one minute, to
allow the initial warm up of the amp. Should
something go wrong inside, the amp turns off and
waits to see what happens next. If nothing
happens, it turns back on. Let me demonstrate how
the delay works (demonstrates).
As
has already been admitted by the experts in the
U.S., England and Japan, these amplifiers, beside
other qualities, have an absolutely unique
capability to reproduce natural bass, although
these are not the DC amplifiers: there are
coupling capacitors between the front end and
output stage...
AM: Which stage of your Soviet
biography do you remember today from the position
of a high-end American designer and LAMM
President and Director of Engineering?
VS: In 1963, I got admitted to
the Lvov Polytechnic University, Department of
Electrophysics, where I majored in solid-state
physics, semiconductors, etc. Graduated in 1968
with a research profession of an
engineer/electrophysicist and designer of
semiconductor devices. Conducted fieldwork and
was working on an M.S. thesis at the Kiev
Institute of Semiconductor Devices of the Academy
of Sciences of the Ukrainian S.S.R. Did research
on electro-luminescence and was involved in
making electro-luminescent devices utilizing
silicon carbide (SiC). Completed a write-up of an
M.S. thesis about 4-5 months prior to the
scheduled defense and was assured of a position
at the Institute upon defending. I had already
started sorting out ideas about my future Ph.D.
dissertation, but in 1968 I got drafted to the
Army and this process was interrupted. (Having
been born in a miserable year of 1945, I belonged
to a generation that was turned into a kind of an
experimental scapegoat on which various social
structures were being tested: first a 7-year
primary schooling, then an 8-year schooling,
special polytechnic education, 11-year primary
schooling...When I was still in college, the Army
service was not required for college
graduates--upon receiving a degree, the graduates
were also conferred a military officer rank which
eliminated the need for them to serve in the
Army. Then, just prior to my graduation, a newly
implemented idiotic Soviet "law" went
into effect making serving in the Army compulsory
for all young graduates...)
When
I came back to Kiev after a two-year military
service I learned that the position I had been
promised before my military service was already
filled. I then returned to Lvov and started
working at the Lvov TV factory
"Electron". In reality, this factory
was a major defense organization. My primary work
there was concerned with closed-circuit TV
systems. Later on we started working on
high-definition TVs--have you ever watched on TV
the Center for Space Flight Control in the town
of Podlipky, seen all those desks with monitors?
That was our design. We also worked on electronic
equipment used for instruction and training of
military and civil pilots and astronauts.
At
one time the factory got a hold of a foreign
TV-system with image scaling that was performed
via optical methods utilizing highly complex
optics. We had been assigned a task of creating
the same system but with a scaling factor of 40
which had to be maintained on the display to keep
up with changing of the television raster and
with the characteristics that had to remain
constant. I started checking around...
AM: This was also a military
order?
VS: Yes. ...I had visited the
Central Aero and Hydrodynamics Institute where
the research was being conducted on a similar
topic. The researchers there explained to me that
such issues could not be resolved by electronic
means (the computers had been a rarity in those
days), only by optical ones. In the Moscow
Television Research Institute, similar systems
had been implemented with scaling of only 5
times, and I needed a 40-times scaling. I took on
this task, maybe because I was too young and did
not realize what was involved in it, but I
finally solved this problem. I quit
"Electron" a couple of years
afterwards, just when various R&D companies
for consumer electronics (and especially audio
electronics) started opening up throughout the
USSR. I turned to the topic I was most interested
in--audio. Worked in the position of a Chief
Design Engineer at the R&D Department of a
large electronics company in the city of Lvov;
The Department was involved in design of consumer
electronics. Some of the projects I oversaw and
participated in were: design and implementation
of various power amplifiers, control and phono
preamplifiers and equalizers for consumer and
professional use; investigation and analysis of
the world audio market; prognosis and formation
of the promising avenues in the development of
consumer audio equipment. "Amphiton"
amplifiers that I had designed and implemented
into production are just an external part of my
job--it is of no interest. At the same time I had
been "making myself" or, in other
words, kept asking myself questions. When I was
unable to answer them, I would embark on a
learning crusade. Learning is a life-long process
for me.
First,
I kept asking the so-called simple and naive
questions. For instance, we have three
amplifiers: one with a 1-2% distortion, another
with a 0.1% distortion and yet another with a
0.001% distortion. All three are heard through a
speaker with, let's say, about 5% of its own
distortion. Theoretically, we shouldn't hear the
difference, just a "signature" of the
speaker. But the paradox is that we hear the
difference between these three amplifiers
perfectly. The answer to this question is not
that easy: it took me a few years to understand
the nature of this effect. There is no place
where you can learn how to make the sound--you
can confirm this with any serious audio designer,
or just any designer who has been able to create
a reasonably wellsounding system. I repeat again
that I have been learning all my life. I had an
opportunity to conduct an enormous number of
experiments or, in other words, check my
theoretical hypotheses and calculations and
narrow the number of topologies capable of
producing good sound until I obtained the desired
results. As strange as it may sound, a big help
in getting the results was my eight-year-long
wait in the Ukraine for an exit visa, although
this period was full of problems related to my
status of a "refusenik" (persecution by
the police and other problems). I had time to
sort out a lot of ideas and results. Based on the
electronic modeling of a human hearing mechanism
and having arrived at a limited number of
electronic topologies that describe our hearing
perception to a greater or lesser degree, I
created a number of working prototypes. I
purposefully created various systems:
solid-state, hybrid, and vacuum tube ones. In the
end, I came to certain conclusions which I have
been able to implement only here, in the U.S. Of
course, it would be impossible for me to do
anything here without years of previous thorough
research--R&D is very expensive here. There
are no grants available for this type of
research, the government is not involved, various
financial groups that sponsor research are not
interested in this type of venture--financially,
this is not the kind of business they would be
investing in. I was able to realize my ideas with
the least amount of expenditures only because
back in Russia I had tremendous experience in
implementing my own designs into production and
their maintenance. I am familiar with an entire
cycle: from an idea to a finished product. (By
the way, back in Russia I was in charge of a
number of very serious research projects and my
job involved finding specific solutions to a
range of problems, from research ideas to their
implementation.) Here, in the U.S., I do
everything myself.
AM: By the way, what do you
think about domestic vendors of parts you use in
your designs? Was it long before you found
vendors? What is the quality of parts?
VS: As I mentioned before, I
became a partner at "Madison Fielding,
Inc." and got to know certain vendors over
time. People don't talk much here, no one teaches
you: if you got yourself into something you have
to handle the consequences on your own. The only
thing that I received assistance with was the
transfer of literature, catalogs, component data
to my home, in order to give me an opportunity to
work out of home when necessary. "Madison
Fielding, Inc." was 50 miles away and I did
not have a car at the time; I had to do all the
commuting by train.
AM: Let's talk about equipment
measurements. There are two points of view on the
subject. First is that measurements do not mean
anything other than (roughtly speaking) quality
of the equipment--we need measurements only to
make sure that everything works properly. Second
is that, at this point, we don't even know what
to measure. Do we need measurements at all?
VS: Of course. The
measurements describe objective processes and
phenomena. In a way, I solved this riddle: I
achieved the correlation between the quality of
sound and formal measurements, or at least those
measurements which are formally present in every
national standard and which are made public by
designers. Of course, when one looks at the
measurement results presented in a graphical way,
one should have at least a basic understanding of
a topology of the measured device and
understanding of which formally measured
parameters are critical to one or the other
specific schematic solution. And it is absolutely
necessary to know the types and values of
feedback (if any) utilized in a given device. In
other words, one should know what to measure.
When a person sees formal measurements of a piece
of equipment--even if he doesn't know anything
about it--he can say a lot about it by looking at
the measurements, as long as he has an
understanding of a subject matter. Usually,
electronic engineers think that audio electronics
is a passing stage between an entry level and
professional electronics, although in reality one
should get into audio electronics after all other
areas of electronics have been mastered.
As a
rule (and I don't know of any exceptions) a
person who is involved in design of at least
high-end audio equipment either has to be a
musician or at least know the sound of a real
orchestra. This is a trivial observation but it
is true.
AM: Let's talk about
assessment of sound quality. In general, there
exist just a few fundamental approaches in this
respect. One of them is quite obvious and
clear--it is when a person compares the
reproduced sound to the one of real instruments
in a natural setting. Another is the one
popularized by Mr. Qvortrup who claims that we
know nothing about a recording process and that
the only way to understand anything is to listen
to an enormous number of records, one after
another, and watch how the difference between
them is revealed. If the difference is
three-dimensional, then the system is
transparent; if it is not, then the system is not.
VS: What if it IS and IS NOT
simultaneously, as so often happens in life? What
are we to do then? (laughing)
AM: But, still, what is your
opinion?
VS: I have already said that a
person should either be a musician or listen to
live music on a regular basis, and to also
remember one thing: that we are the perceivers
and that the mystery is beyond us. The only thing
we can do (using a servo system for analogy) is
to choose a reference point based on live music
and compare everything else to this point...Then,
this question does not even arise.
This
is an old issue, an issue of inner culture and
education. We live in this world, we have
certain musical instruments, we know certain
music by certain composers--and we are limited by
this information. Why, then imagine supernatural
problems and put a heavy burden on oneself? It is
important to accumulate "baggage"
first--that is, to know how the real orchestra
sounds. The rest is a question of classification
and creation of certain hierarchical gradations.
The problem of sound quality assessment is almost
completely solved in the first 10-15 seconds of
listening at the intuitive level. Any experienced
listener knows that the first minute of listening
is the decisive one, and the rest is simply
putting the perceptions of that one minute into
words; it becomes a question of terminology and
method of communication.
As
human beings, we perceive sound on various
levels: conscious as well as subconscious (or
intuitive). We perceive sound not just with our
ears, but with the whole body. I had been
seriously involved in sound perception issues and
developed a very specific criteria for myself in
this regard.
We
have to listen only to the music; the equipment
should move into the background, disappear. We
must listen to the music--I think, this is the
only reasonable approach. Right now, I am not
talking about the questions of objective
knowledge and so forth.
From
my point of view, the ideal review is when an
audio reviewer has nothing to say. Quite
naturally, such reviews do not get published(!),
as the reviewers must earn their bread, and the
readers must get their entertainment; but the
ideal conclusion should be: "I have nothing
to say..."
AM: What are your plans for
the near future? What new equipment are you
working on?
VS: In the of LAMM AUDIO
LABORATORY® products (solid-state and hybrid
line) I have completed the design of a dual-mono
power amplifier model DM1, with a retail price of
approx. $9,700. Aesthetically it will look very
similar to the M1.1, with a wider front panel (19
inches in width) and two heatsinks on each side.
This is a highly biased A/AB design; the topology
and physical design are similar to that of the
M2.1, with an exception of a tube stage with
accompanying elements--filament voltage regulator
and so forth that are quite expensive--which will
be replaced with a high voltage MOSFET. In other
words, the amplifier will be a pure solid-state
design. As I said, the design has been completed,
a prototype built, and what's left is to make the
PCBs and start production. I might make a small
preproduction run at some point in the near
future. Based on the DM1, I will have a more
affordable stereo amplifier, at about 60 Watts
per channel; it will be the size of my
preamplifier L1. This stereo amp will also
feature high-speed MOSFET transistors in the
output stage with no overall feedback, and
topology and design philosophy similar to my
other designs. It will have the same
"recognizable" LAMM sound. This model
will retail at approximately $4,000-4,500. I will
also have a simplified version of the L1 that
will retail at $2,900-3,100. With this, I will
draw a line to my solid-state designs for the
near future.
I am
a "vacuum tube" man by nature. There is
no limit to what can be done. I even established
a new company LAMM INDUSTRIES, INC. which
manufactures vacuum tube equipment. I have
practically completed the design of a
single-ended 8-Watt amplifier on 300B--this is
more or less tribute to the latest fashion. I
also have a design of a push-pull amplifier on
300B, operating in class A, but I would rather
not start its production just yet, as I plan to
start the production of an 18-Watt single-ended
amplifier model ML2 shortly. And the crowning
touch of all will be a single-ended amplifier on
a Russian tube GM-70.
I
have practically completed the design of two more
models of more affordable power amps on vacuum
tube 6550 (which is, by the way, from
"Svetlana"); this tube has gained a
fairly good reputation, especially its latest
modifications. One design is a 70-75-Watt stereo
amplifier, with two 6550 in each channel; and the
other is a 150-Watt monoblock, with four or six
6550 tubes per channel. I will also have quite an
original (probably even unique) model of a vacuum
tube preamplifier.
To
sum up, I will have three single-ended models:
8-Watt on 300B; 18-Watt model ML2, and 30-35-Watt
one on GM-70. Right now, I have completed a
design of quite a fascinating preamplifier which,
I think, will become the reference (for some time
to come) in terms of sound quality. By the way,
it will have a Russian tube 6N6P that has certain
interesting characteristics.
What
I'm showing you now is a photo with a phono
preamp--it is practtically ready, I'm just
working out the details. It will be quite an
expensive piece, and there is a relatively
limited, although steady, market for such
equipment. As strange as it may seem, in spite of
the cornucopia of phono stage models available on
the market (solid-state as well as vacuum tube
ones), serious audiophiles still feel the lack of
a real phono stage that would be practically
neutral and satisfy all requirements. Many of my
customers and dealers keep asking when I would
have such a phono stage.
It is
very difficult to do everything simultaneously.
The key is in increasing our sales. Export is
important, including Russia, although up until
recently I have been very skeptical about trying
that. However, I now begin to think that Russia
might become an interesting market in terms of
audio. One of my goals is to get to the Russian
and Ukrainian markets and other regions which had
become independent. It is only a question of time
when all this outrageous money and foolishness
that are characteristic of today's Russia
transform into respectability. From the children
of all those who had at one point been
slaughteered, shot, put to prison and destroyed,
there will emerge (of course, under a different
guise) a new class of people. Natural sorting out
always takes place. Equal civil rights do not
presuppose equal individual capabilities; each
individual is made in a certain way, each
individual needs certain things--it has always
been like that and will always stay like that.
I'm
involved more in technical aspects than in
political issues. When one leaves Russia, the
unhealthy interest in the so-called politics goes
away, and other aspects of existence come to the
fore, those aspects which did not even touch our
paranoidal mind at the time when all of us had
been deprived of our individuality.
Many
dealers and designers ask me what I think of
digital technolgoy and whether my company will
every go digital. My answer is very simple: this
field is not a terra incognita for me but I
consider myself quite an ordinary engineer in it.
I always conform to my philosophy of not becoming
involved in things which many people are able to
do. In order for me to achieve a level of art in
digital technology I have to live one more life.
First of all, it is a little too late to get into
this, and, second, my heart is not in a digital
domain. I have always tried to do what I can do
best and have paid a tremendous price to be left
alone in order to do what I love.
I'm
an "analog" man by nature and my mind
is "analog." I would not want to
exchange the best digital system for the best
analog system. Simple listening tests of digital
and analog systems of comparable level make clear
that there is nothing to talk about (fortunately,
there is a great number of high quality LPs, and
CD reissues of those LPs that make such tests
possible). Digital technology by itself is not so
bad; it is convenient in terms of working with
recorded material (remote control operation, ease
of choosing any track on a CD, choice of various
programs and so forth), but if you want to
"fly away" and have serious listening,
there are no two opinions about it.
The
market situation here is very interesting: there
are many worthwhile things, along with quite a
number of cheap promotional tricks. Take, for
instance, an amplifier of class A operation. The
fact that the designer decides to create an
amplifier operating in pure class A (especially
high power class A amp: over 50 Watts)
automatically presupposes that it will be fairly
expensive just from the standpoint of physical
implementation. I leave out the issues of sound
quality--everything depends on who and how will
be designing such an amplifier. Here, in the
States, the fact that the audiophiles-at-large
know only that class A is a good thing gets
abused: all too often anything slighly warmer
than liquid nitrogen is declared class A. Because
the consumers know that class A operation is a
good thing, certain manufacturers, in order to
promote their products, use the term "class
A" to describe amplifiers operating in class
AB or even B. As a result, a problem arises with
price formation for the amplifiers that actually
work in class A. Take, for instance,my ML1
amplifier. Using audiophile terminology, its
class of operation can be described as highly
biased A/AB. However, I could quite comfortably
label it an amplifier operating in pure class A.
And, of course, it would be much easier to sell
it then. However, this is plain cheating. In
reality, should the ML1 be a pure class A
amplifier (the same 80-90 watts of the output
power with a frequency range of 20-20,000Hz) done
honestly and properly, the cost of its building
would rise significantly. But in this case, and
it would be very, very difficult to sell it,
although the jump in sound quality would be
significant. And so, we are back at square one.
When
a piece of equipment is designed properly,
so to speak, then formal technical data describes
and characterizes it quite comprehensively. You
know, for instance, that single-ended amplifiers
(especially those on vacuum triodes) have a weak
spot: roll-off at the low and high frequencies
because of an output transformer. For the absolute
majority of single-ended tube
amplifiers, frequency response characteristic is
specified at the output power of 1 watt. What's
going on with the frequency response at the
nominal power is a "black hole"...--
and this is one of the most important parameters
of any amplifier, and especially single-ended
one. My single-ended amplifier model ML2
maintains a frequency response from approximately
16-20Hz to 60-80KHz (exact values will be
specified when production begins) at the nominal
power. And I guarantee the constancy of
harmonic residue shape at all frequencies in the
range of 20Hz-20KHz (again, at the nominal
output power!). There hasn't been
anything like that on the market among the
single-ended amplifiers.
M1.2
Reference
