The problem with all the comparisons to 8 watt SETs...
is that "all else is not equal".
Philefread,
Your 8 watt SET amp IS "loafing".
That 300B tube is capable of more than just 8 watts.
That tube can be used in amps putting out on the order
of 40 to 50 watts. For example, Cary uses it in a
15 watt SET power amp:
http://www.caryaudio.com/products/audio/cad300se.shtml
Therefore, in the context of my discussion on linearity -
the tube is not being taxed at all.
Your 300B SET amp is not limited by the 300B tube - it is
limited by the power supply.
The designer / manufacturer of your amp put in a low power
power supply - so you can't push the tube to anywhere near
its capability.
The manufacturer forced the tube into "loafing" by pairing
it with a relatively "anemic" power supply.
Your speakers may be relatively flat - but they don't have
an absolutely flat 8 ohm impedance. Do your speakers have
voice coils? Voice coils are inductors - and the impedance
of an inductor is not flat - it varies with frequency.
Dr. Gregory Greenman
Physicist |
Raquel,
The problem with your argument is that in amps in which
there are more than a single output device; the output
devices are in parallel.
No - the signal is not going through more circuitry - all
the electrons see the same amount of circuitry. It's just
that all electrons don't see the same components.
If you have a tube amp with 2 paralleled 300B tubes; all the
electrons saw a single 300B output tube - just not the
same one - 50% saw one tube, 50% saw the other.
The comparison at Singer is worthless. A VERY BIG factor
is the size of the room. One was smaller, one was larger.
The rooms therefore had different resonant frequencies -
and Lord only knows how that affected the sound.
You can't conclude anything even remotely meaningful from
two different rooms, with two different speakers, with
two different amps...
Dr. Gregory Greenman
Physicist |
Raquel,
Here's the web page of someone that has actually tested
the tubes they're dealing with:
http://jcverdiervalvevinyl.online.fr/Kinky300/300Binfo.html
The quality of the tubes varies - which is true of
every manufactured component. However, it is especially
true of hand-assembled devices like vacuum tubes.
However, one can see that even in the tubes labeled "perfect",
there is a degree of non-linearity. The plate characteristics
curves are not perfectly straight - and they are not
perfectly spaced.
The author is correct that they are "perfect" - but
perfection only goes so far with a hand-assembled
1935 designed electronic device.
The 300B is not bad for a tube designed in 1935.
In fact, it's a rather impressive tube for a
device that did not benefit from modern computer
modelling, and modern electronics manufacturing
techniques.
However, one can produce devices with better linearity.
Dr. Gregory Greenman
Physicist |
Raquel,
In addition to all the differences between the two systems
that you note - the rooms had different room modes - they
accentuated / deaccentuated different frequencies. That's
probably the biggest factor making the comparison of the two
systems invalid.
When comparing audio equipment - at least stick to the same
or similar rooms. You'd be surprised how much the room
alters the sound.
As far as you thinking of me with mis-matched socks, that's
just your own prejudice showing in thinking of me in a
stereotypical fashion; a stereotype that has little backing
in truth. I'm not a stereotypical scientist; I don't know
any who are.
However, I am enough of a scientist to know that all the
fine points of audio have physical explanations. There's
no "magic" at work in tne audio field.
Dr. Gregory Greenman
Physicist |
Raquel,
Again, you ignore the science that all else being equal,
the higher power amp is more linear - that's just a FACT
of science.
I've read John Atkinson's pg 1 Stereophile editorial. I'm
not commenting on blind testing at all. I'm talking PHYSICS!
You may have heard a lot of equipment since 1977, but I've
both heard and measured equipment since 1977, and I know
the physical principles on which audio is based.
You can't get around the fact that the less you push the
active devices - be they tubes or transistors - the better.
Real tubes and real transistors are not linear.
The problem with audiophiles that only listen and never
measure is that they confuse what they like for accuracy.
I've experimented with audiophile friends, where I distort
the signal on purpose - rolling off frequencies, altering
time constants, etc. and the audiophile prefers it!
I believe a lot of audiophiles prefer music that has had
the harsh "edges" removed by some wimpy amp. You may
prefer it - but it is not accurate.
Real musical instruments have these "edges".
From your listening experience, though vast, you can only
say what you like - you can't say what is accurate.
Accuracy comes in making comparisons of REAL instruments
with recordings of THAT instrument - which is what I've done.
You can't make a blanket statement that low power amps
are "cleaner". The only thing they've "cleaned up" are
real characteristics of real instruments.
Dr. Gregory Greenman
Physicist
|
A1126lin,
The reason for the seeking higher power is "linearity".
If amplifiers were perfectly linear - that is their
characteristics were independent of the magnitude of their
output, then Gmood1 and Nrchy above would be correct -
there would be no advantage to having more power than
you need.
However, amplifiers are NOT perfectly linear. They are only
approximately linear. One way to stay within the most
linear regime of the amplifier's operating characteristics
is to use a small fraction of the amplifier's ultimate
power. That is the more the amplifier "loafs" - the
better.
Who would you trust more with lifting a prized breakable
possesion weighing 40 lbs - someone that can lift 60 lbs
or a longshoreman that routine lifts over 100 lbs?
The less the load "taxes" the amplifier - the better.
Now does high power necessarily equate to better quality?
Certainly NOT - Thorman is correct there.
The design of the amplifier, the quality of the components,
etc; all are factors affecting the quality. However, all
else being equal - the more powerful the amp - the more
you stay in the linear regime of the tubes or transistors
that make the amp work.
You also have to be cognizant of the fact that the response
to power is logarithmic. That is 300 watts is only a
1.76 dB increase over 200 watts, and only 4.78 dB over
100 watt power level.
So what may look like a large increase on a linear scale -
is actually only a small increase on a logarithmic scale.
Dr. Gregory Greenman
Physicist |
S7horton,
Let's take the comparison of the Classe CA301 at 300 watts
and the Classe CA201 at 200 watts.
If the design of the amps is essentially the same, and Classe
selects components with the same quality; but it just puts
more transistors in the CA301 than it does the CA201; then
yes.
For the same load, the transistors in the CA301 will each
be providing less current than their counterparts in the
CA201. Therefore, because the transistors in the CA201
are putting out more current - they are farther out on
the transistor's response curve. Depending how linear
that curve is - indicates how much less accurate the CA201
will be than the CA301.
If the curve was perfectly linear - then there would be
no difference. But real transistors never have perfectly
linear curves. The CA201's transistors are extending
farther out on the curve than the ones in the CA301,
because the CA201 has fewer transistors to carry the same
load.
The difference may be marginal, if the transistor's
characteric curves are fairly linear; but there is a
difference because the curves for real transistors are
never perfectly linear.
Dr. Gregory Greenman
Physicist |
Raquel,
Yes - I read your Singer anecdote - and noted the qualification.
However, the conditions of the comparison made that
comparison absolutely WORTHLESS.
So worthless - that it shouldn't even be brought up -
qualifications notwithstanding.
I reject your contention that there is one "way to go"
in audio system design - efficient speakers and relatively
low power amps.
Some well executed speaker designs, for example ribbons;
are inherently low efficiency and require high currents.
You can only make a permanent magnet so strong. A ribbon
driver has inherently a single turn in its "voice coil" -
so unlike dynamic drivers where the efficiency can be
increased by putting more turns in the voice coil - a
ribbon can't take advantage of that technique. With a
limited strength magnet, and a single turn - the only way
to get more output is the third term in the equation for
the Lorentz force - the current. The impedance must be
lowered to obtain more current. However, that means a
more powerful amp.
That approach is every bit as well considered as your
approach.
As far as the 300B being linear - ARE YOU KIDDING!!
Have you ever put a 300B on a test bench to determine
the linearity? I have - and I would NOT say that they
are champs in the linearity department.
One can make an amp that sounds very nice to those that
like the sound of the 300B, but please don't say it is
anything special when it comes to linearity.
Dr. Gregory Greenman
Physicist |
Eldartford,
Very good - I agree with you 100%
All else being equal [ which is the key ], the more
powerful amp is more linear.
Some audio systems - some speakers - require the higher
power amps; like the ribbons of a Magneplanar.
Some speakers are very efficient and won't tax even a low
power amp - thus keeping it in the more linear regime.
I get really irked by people that say there is ONE WAY
[ usually their way ] to accomplish a given task.
If that were true, we'd all be driving the SAME car,
because there would be only one BEST car, watching the
SAME TV, because there would be only one BEST TV....
There are a whole host of different philosophies. Any
engineer will tell you that engineering is about making
compromises and trade-offs. You trade one set of advantages
for a set of disadvantages. What choices one makes is
determined by what they value.
As the old saying goes - "That's why they make chocolate
and vanilla!" Chocolate is NOT necessarily the one BEST
choice in ice cream.
Dr. Gregory Greenman
Physicist |
Raquel,
I'm well aware that Sound by Singer is a well respected
audio dealer - and they set up their rooms properly.
What you fail to appreciate is that you can't take the
room out of the equation. No matter how you treat the
walls of a room - the walls still reflect sound.
My comments are not ill-reasoned. Solving the wave
equation that dictates the propagation of sound in air
is one of my specialties.
The propagation of sound in a room is dictated by the
wave equation; a second order partial differential equation.
Mathematics requires that one provide "boundary conditions"
in order for the problem to be well posed.
The walls of the room provide those boundary conditions.
No matter how you treat the walls - they still reflect
sound to a degree. Because of the reflection of sound from
the walls; you get "room modes".
For example, if the room is 20 feet in some linear
dimension - length or width - then there are room modes
at frequencies of 26.6 Hz and 53.2 Hz corresponding to
a half-wavelength and a full-wavelength fitting that
dimension, respectively.
This gives you "standing waves" at those frequencies -
the music at 26.6 Hz and 53.2 Hz will be accentuated.
If you move to a different room, with different dimensions,
then different frequencies are accentuated.
There is no way that the people at Sound by Singer can
get around this. It is just a simple fact of life when
it comes to audio, that is dictated by the Physics of
sound travelling in air.
Because you auditioned two different systems, in two
rooms that you stated were different in size; then the
rooms were accentuating different frequencies and thus
you would have heard differences even if you were listening
to the exact same audio setup.
Because you don't know what the room is doing to the
sound, any comparisons where the room acoustics are
different is invalid.
Dr. Gregory Greenman
Physicist |
