[Dave]

A acquaintance of mine is wanting to upgrade his amplifier, an EL34-based
tube amp wired in ultralinear configuration. I'm considering buying it, but
am unsure if this amp's 20 wpc is enough to drive my Boston Acoustics T-830
speakers. Right now they're driven with a 50 wpc Harmon Kardon vintage
solid state amp, and that setup can be played LOUD. If I purchased the tube
amplifier, I'd likely want to rewire it in triode configuration which would
halve my power output to around 10 wpc.

Does anyone know the efficiency specs for the T-830 speakers, or any of the
T-series BA loudspeakers from the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Thanks

Dave S.

[Codifus]

On Oct 23, 7:51 pm, "Dave" wrote:
A acquaintance of mine is wanting to upgrade his amplifier, an EL34-based
tube amp wired in ultralinear configuration. I'm considering buying it, but
am unsure if this amp's 20 wpc is enough to drive my Boston Acoustics T-830
speakers. Right now they're driven with a 50 wpc Harmon Kardon vintage
solid state amp, and that setup can be played LOUD. If I purchased the tube
amplifier, I'd likely want to rewire it in triode configuration which would
halve my power output to around 10 wpc.

Does anyone know the efficiency specs for the T-830 speakers, or any of the
T-series BA loudspeakers from the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Thanks

Dave S.

Don't really need to know the efficiency. Going from 50 watts down to
10 is a gigantic loss of power. Every doubling of power raises the
volume output of a system by only about 3 db. Your tube amp will be
cranking at full rated power and be pushing the speakers at just above
conversation level. And since it's running at full power, distortion
will occur more readily for any slight increase in dynamics of the
music.

If you had some very effiicent horn speakers like Klipsh, then maybe.
But with regular cone speakers?

Don't do it.

CD

[Peter Wieck]

On Oct 23, 6:51 pm, "Dave" wrote:
A acquaintance of mine is wanting to upgrade his amplifier, an EL34-based
tube amp wired in ultralinear configuration. I'm considering buying it, but
am unsure if this amp's 20 wpc is enough to drive my Boston Acoustics T-830
speakers. Right now they're driven with a 50 wpc Harmon Kardon vintage
solid state amp, and that setup can be played LOUD. If I purchased the tube
amplifier, I'd likely want to rewire it in triode configuration which would
halve my power output to around 10 wpc.

Does anyone know the efficiency specs for the T-830 speakers, or any of the
T-series BA loudspeakers from the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Thanks

Dave S.

Hmmm.....

OK, let's do some basic math that is irrespective of speaker
efficiency. Assume that BA speakers are relatively inefficient. So,
let's figure say.... 86dB @ 1 watt. So, a 20-watt amp will be capable
of giving you 99dB without clipping. A 40-watt amp will give you
~102dB, and a 50-watt amp will not make any perceptable difference to
that 102.

A 50-watt amp will be capable of giving you ~102/3 dB without
clipping. 3dB is a very small increment relative to all things, one
that you can definitely hear with normal hearing but not earth-
shattering.

10 x the power = 2 x the volume perceived. So, 1 watt gives 86dB, 10
watts give 96dB, 100 watts gives 106dB.

The calculations for the dB relationships a for a 10 to one
relationship, the log of 10 is 1, and ten times 1 is 10. For the 2 to
one relationship, the log of 2 is 0.3, and 10 times that is 3.

Put another way, after 20 watts, the next truly meaningful increment
as to headroom under normal conditions and normal listening levels is
roughly 100 watts. The value of headroom decreases with the efficiency
of the speaker. And it increases with the peak-to-average levels of
the source material.

Do a search on "watts to decibels" and you will get much useful
information. But it comes down to the fact that there is little actual
difference in capacity between amps say.... over 10 watts and under 50
watts under most conditions.

Peter Wieck
Wyncote, PA

[[email protected]]

On Oct 23, 11:21 pm, codifus wrote:
On Oct 23, 7:51 pm, "Dave" wrote:

A acquaintance of mine is wanting to upgrade his amplifier,
an EL34-based tube amp wired in ultralinear configuration.
I'm considering buying it, but am unsure if this amp's 20
wpc is enough to drive my Boston Acoustics T-830
speakers. Right now they're driven with a 50 wpc Harmon
Kardon vintage solid state amp, and that setup can be played
LOUD. If I purchased the tube amplifier, I'd likely want to
rewire it in triode configuration which would halve my power
output to around 10 wpc.
Does anyone know the efficiency specs for the T-830
speakers, or any of the T-series BA loudspeakers from
the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Don't really need to know the efficiency.

Well, yes, you kinda do, especially given the assertion
you make later on.

Going from 50 watts down to
10 is a gigantic loss of power. Every doubling of power raises the
volume output of a system by only about 3 db.

Okay, then let's work that back. since

dB = 10 log (P1/P2)

an P1 = 50 and P2 = 10, then the loss in ultimate level,
all other things being equal, would mean a loss of 7 dB.

Is 7 dB a "gigantic loss in power?"

Your tube amp will be cranking at full rated power and
be pushing the speakers at just above conversation
level.

Now, indeed, you make an assertion that assumes you
know the efficiency of the speakers. Or, from your assertion,
we can derive the efficiency.

Normal "conversation levels" are considered in the realm of
60-70 dB SPL (cf Kinsler, Frey, also Beranek, Olson, and
confirmed by my own measurements). Let's assume that
by "just above" you mean 3 dB, or twice the acoustic power.

That would mean that you're, in effect, asserting the gentleman's
speakers have an efficiency in the realm of around 65-70 dB
SPL/watt, making them, far and away, the LEAST efficient
speakers I have ever encountered.

Instead, knowing the types of drivers used, we can probably
reasonably assume a broad-band efficiency in the realm of
86 dB SPL @ 1 watt, fully 20 dB MORE efficient than is
implicit in your claim, that would suggest that the system
as he describes it should be capable of playing AT LEAST
30 dB LOUDER than "conversation levels," putting it in
the realm of 95-100 dB before there's an issue.

But with regular cone speakers?

Don't do it.

Since I would assert with reasonable confidence that
MOST of the time he's listening with the amplifier coasting
along at less than a watt, that there is absolutely NO reason
whatsoever NOT to try it.

[Dave]

"Peter Wieck" wrote in message
...

Opinion #1:

information. But it comes down to the fact that there is little actual
difference in capacity between amps say.... over 10 watts and under 50
watts under most conditions.

Opinion #2

If you had some very effiicent horn speakers like Klipsh, then maybe.
But with regular cone speakers?

Don't do it.


I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Dave S.

[graham]

"Dave" wrote in message
...
A acquaintance of mine is wanting to upgrade his amplifier, an EL34-based
tube amp wired in ultralinear configuration. I'm considering buying it,
but
am unsure if this amp's 20 wpc is enough to drive my Boston Acoustics
T-830
speakers. Right now they're driven with a 50 wpc Harmon Kardon vintage
solid state amp, and that setup can be played LOUD. If I purchased the
tube
amplifier, I'd likely want to rewire it in triode configuration which
would
halve my power output to around 10 wpc.

Does anyone know the efficiency specs for the T-830 speakers, or any of
the
T-series BA loudspeakers from the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Thanks

Dave S.

1993 Audio Annual Equipment Buyer's Guide

Boston Acoustics T-830:

Frequency Response: 40Hz - 20kHz +/- 3dB
SPL: 1 watt/1 meter / 88dB
Recommended Min. amp power: 15 watts per channel

[Walt]

codifus wrote:

Don't really need to know the efficiency. Going from 50 watts down to
10 is a gigantic loss of power. Every doubling of power raises the
volume output of a system by only about 3 db.

Uh huh. So a 50 watt amp will be capable of 7db more output than a 10
watt amp.

Is 7 db SPL too much to give up?

Also, since the lower power amp is a tube amp, it won't sound as nasty
if it's overdriven, so you can push it harder than a 10 watt solid state
amp.

//Walt

[Randy Yates]

Walt writes:

codifus wrote:

Don't really need to know the efficiency. Going from 50 watts down to
10 is a gigantic loss of power. Every doubling of power raises the
volume output of a system by only about 3 db.

Uh huh. So a 50 watt amp will be capable of 7db more output than a 10
watt amp.

Is 7 db SPL too much to give up?

Also, since the lower power amp is a tube amp, it won't sound as nasty
if it's overdriven, so you can push it harder than a 10 watt solid
state amp.

The problem is that 7 dB is in a gray area. It's a pretty big drop, but
it's not THAT big. How loud was "loud" in the original setup? If it was
just "barely loud" then 7 dB would probably drop that to "not loud."

In my opinion, 10 wpc on 88 dB SPL speakers is wimpie wimpie wimpie.
But then I'm a Klipschorn owner...
--
% Randy Yates % "How's life on earth?
%% Fuquay-Varina, NC % ... What is it worth?"
%%% 919-577-9882 % 'Mission (A World Record)',
%%%% % *A New World Record*, ELO
http://www.digitalsignallabs.com

[Peter Wieck]

On Oct 24, 6:51 pm, "Dave" wrote:

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Dave:

Far be it from me to render a "third" opinion. Rather I would suggest
that you check an independent source or two on the relationship
between watts and decibels as it applies to your speakers. So, you
will get 88dB at 1 watt at 1 meter from your BA speakers.

88dB is not LOUD, but it is substantially above conversational level,
roughly 2x louder.

1 watt = 88dB
10 watts = 98dB
20 watts = 101dB
50 watts = 104/5 dB

All other things being equal on those speakers.

You will not miss the increment between 20 and 50 watts for two
reasons one a bit esoteric, the other very practical:

a) Tube amps clip softly. So, if you are listening at an average of
one watt and your peak-to-average on your source is 20dB (well-
recorded classical music will reach these standards), you will clip at
the peaks but do no substantial damage to your speakers or the amp if
clipping does not exceed... well... let's use an arbitrary 20% of the
time - and even then damage is quite unlikely as output transformers
will not pass DC.

b) You will clip almost equally with a 50 watt amp with the same P/A
source. However, a "vintage SS amp" *will* pass DC at clipping - which
*will* damage the speakers eventually or immediately depending on the
levels at the moment.

So, you pays you money, you takes you chances. I would be far more
skeptical of you going to UL at 10 watts and being satisfied unless
you listen at pretty low levels and/or you listen to mostly compressed
sources (P/A 10dB) if only because your speakers are energy pigs.

As it happens, I really prefer piggish speakers personally, I have a
pair of AR3as which are rated at 86dB, but then I feed them with a 300
watt solid-state amp (@ 4ohms), so headroom is not an issue. My more
efficient speakers are perfectly happy being fed either from my Scott
LK150, Dynaco ST-70 or ST-35 being 75, 35 and 15 wpc respectively, and
tube amps.

Peter Wieck
Wyncote, PA

[[email protected]]

On Oct 24, 6:51 pm, "Dave" wrote:
"Peter Wieck" wrote in message

...

Opinion #1:

information. But it comes down to the fact that there is little actual
difference in capacity between amps say.... over 10 watts and under 50
watts under most conditions.

Opinion #2

If you had some very effiicent horn speakers like Klipsh, then maybe.
But with regular cone speakers?

Don't do it.

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Well, there are opinions and there are opinions.

In the realm of physics, not all opinions are created equal.
The person who made the claims:

"Don't really need to know the efficiency"

"Your tube amp will be cranking at full rated power and
be pushing the speakers at just above conversation level."

Had no technical justification for these rather
extraordinary claims. In addition, he provided numbers
which contradicted his own claims.

If you try it, and you're sane about it, meaning that
if the music starts to sound too loud and distorted,
it IS too loud and distorted, then the risk is VERY small.

Yes, there are opinion. and there are physical facts.

The physical facts are

1. there's a difference of 7 dB between the rated power
of the two amplifiers.

2. At normal to moderately loud listening levels, with
speakers with an 88 dB@1W sensitivity, you're no
where near requiring 10 watts, much less 50 watts.

3. With only 10 watts per speaker, not only is the
smaller amp capable of playing at level far above
"conversational levels", it's capable of playing well
at levels that would make conversation difficult,
e.g., SPL's in the mid 90 dB realm or better in a
typical domestic living situation in a room of 50 m^3
(1300 ft^3) sitting about 3m (9 ft) away from the
speakers.

4. Boston Acoustics stated the speaker required 15
watts minimum power. That's only 1.8 dB more than
10 watts. 1.8 dB is NOT a tremendous difference, and
could pass as less than a just noticeable difference
under uncontrolled, informal, recreational listening.

There are opinions, and there are physical facts. Physical
facts trump opinions pretty much all the time, unless those
opinions are well grounded in the facts.

[Codifus]

On Oct 24, 6:50 pm, wrote:

Since I would assert with reasonable confidence that
MOST of the time he's listening with the amplifier coasting
along at less than a watt, that there is absolutely NO reason
whatsoever NOT to try it.

The original poster did say that his setup "can be played LOUD." This
would tell me that the user LIKEs that capability.
So wherever your "reasonable assertion of amplifier coasting" comes
from, I don't know. 1 watt? Warming up, maybe.

Forgive me for not calculating this scenario with the accuracy of a
NASA space engineer. You made your point. The 10 watt amp will drive
his speakers. My concern, which I thought may be his as well, was if
it could drive his speakers hard. It won't. The system may sound nicer
on the tubes, but he WILL miss the power.

The Poster Walt brings up a good point in that tubes distort more
pleasantly than solid state. That may help.

CD

[Walt]

Dave wrote:

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

So, a ten watt amp will be capable of driving the speakers to 98 db SPL
at one meter.

A 50 watt amp, which has 7db more output than the ten watter, will be
capable of of driving the speakers to 105 db SPL at one meter.

Now, you're unlikely to be listening at a distance of one meter, so
you'd need to compensate for your actual listening distance. In an
anechoic chamber, you'd drop the db SPL 6db for each doubling of
distance. In a room with reflections it's more complicated, and without
knowing more about your listing room I can't tell you much beyond this:

For moderate listening levels in a small to medium sized room, no
problem. For rock concert levels or in a big room, 10 watts ain't enough.

Why not just wire the darn thing up and see what you think?

//Walt

[Arny Krueger]

"Peter Wieck" wrote in message
...
On Oct 24, 6:51 pm, "Dave" wrote:

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Dave:

Far be it from me to render a "third" opinion. Rather I would suggest
that you check an independent source or two on the relationship
between watts and decibels as it applies to your speakers. So, you
will get 88dB at 1 watt at 1 meter from your BA speakers.

88dB is not LOUD, but it is substantially above conversational level,
roughly 2x louder.

1 watt = 88dB
10 watts = 98dB
20 watts = 101dB
50 watts = 104/5 dB

All other things being equal on those speakers.

You will not miss the increment between 20 and 50 watts for two
reasons one a bit esoteric, the other very practical:

a) Tube amps clip softly.

False claim. Tube amps plenty sharply if they have enough loop feedback to
be usable for hi fi purposes.

Some PR mileage has been made out of the fact that tubed amps often have
relatively poorly-regulated power supplies, but once higher-voltage
transistors became more available, lots of SS amps with relatively
poorly-regulated power supplies have been built and sold.

[Peter Wieck]

On Oct 26, 5:27 pm, "Arny Krueger" wrote:
"Peter Wieck" wrote in message

...





On Oct 24, 6:51 pm, "Dave" wrote:

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Dave:

Far be it from me to render a "third" opinion. Rather I would suggest
that you check an independent source or two on the relationship
between watts and decibels as it applies to your speakers. So, you
will get 88dB at 1 watt at 1 meter from your BA speakers.

88dB is not LOUD, but it is substantially above conversational level,
roughly 2x louder.

1 watt = 88dB
10 watts = 98dB
20 watts = 101dB
50 watts = 104/5 dB

All other things being equal on those speakers.

You will not miss the increment between 20 and 50 watts for two
reasons one a bit esoteric, the other very practical:

a) Tube amps clip softly.

False claim. Tube amps plenty sharply if they have enough loop feedback to
be usable for hi fi purposes.

Some PR mileage has been made out of the fact that tubed amps often have
relatively poorly-regulated power supplies, but once higher-voltage
transistors became more available, lots of SS amps with relatively
poorly-regulated power supplies have been built and sold.- Hide quoted text -

Arnie:

Once again, as this is a regulated forum, I will refrain from a
personalized response.

Tube amps, at least those that use output tranformers do not pass DC
to the speakers. Hence, they clip - SOFTLY - as compared to many/most
vintage SS amps of the same/slightly greater output capacity. And,
hence, they are less likely to damage speakers (or themselves) if
driven to clipping, assuming reasonable build quality in either case.

That would be the point. Headroom is a far more critical issue when
discussing SS amps than it is with tube amps due to their very
different natures. A 50-watt solid-state amp, therefore, is not
substantially more powerful in any meaningful way than a 20-watt Tube
amp.

Peter Wieck
Wyncote, PA

[Greg Wormald]

snip
So, you
will get 88dB at 1 watt at 1 meter from your BA speakers.

88dB is not LOUD, but it is substantially above conversational level,
roughly 2x louder.

1 watt = 88dB
10 watts = 98dB
20 watts = 101dB
50 watts = 104/5 dB

All other things being equal on those speakers.

My understanding (and I'm always willing to keep learning) is that two
speakers (stereo) will provide 3db more volume than one if both are
driven to the same level at the same time. So this reduces the power
needed a bit.

We must also take into account that the 1 watt = 88dB does not account
for musical peaks, which may be up to 30 or more dB above the average
level and will therefore require much more power. IMO this is one reason
that very high power amps often sound better than low power ones of the
same quality.

Of course all calculations must include room size and absorbtion as
well, so at best without real room data, it is all a guess.

In the end, if it works, it works. Give it a try and see whether you
like it.

Greg

[Sonnova]

On Fri, 26 Oct 2007 15:27:39 -0700, Arny Krueger wrote
(in article ):

"Peter Wieck" wrote in message
...
On Oct 24, 6:51 pm, "Dave" wrote:

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.

Dave:

Far be it from me to render a "third" opinion. Rather I would suggest
that you check an independent source or two on the relationship
between watts and decibels as it applies to your speakers. So, you
will get 88dB at 1 watt at 1 meter from your BA speakers.

88dB is not LOUD, but it is substantially above conversational level,
roughly 2x louder.

1 watt = 88dB
10 watts = 98dB
20 watts = 101dB
50 watts = 104/5 dB

All other things being equal on those speakers.

You will not miss the increment between 20 and 50 watts for two
reasons one a bit esoteric, the other very practical:

a) Tube amps clip softly.

False claim. Tube amps plenty sharply if they have enough loop feedback to
be usable for hi fi purposes.

I was taught in engineering school that tubes produced mostly even-order
harmonic distortion when they clipped and that the clipped waveform didn't
look as much like a square wave as does a transistor which, when fully "on"
acts like a switch and produces lots of odd-order distortion. Perhaps, since
even-order harmonic distortion is easier on the ear and more consonant with
music, that this is the source of the term "soft-clipping". Of course, driven
hard into clipping, no amplifying device is listenable, no matter what type
of distortion is being produced but momentary peaks are another matter.

Some PR mileage has been made out of the fact that tubed amps often have
relatively poorly-regulated power supplies, but once higher-voltage
transistors became more available, lots of SS amps with relatively
poorly-regulated power supplies have been built and sold.

[Codifus]

wrote:
On Oct 24, 6:51 pm, "Dave" wrote:

"Peter Wieck" wrote in message

...

Opinion #1:


information. But it comes down to the fact that there is little actual
difference in capacity between amps say.... over 10 watts and under 50
watts under most conditions.

Opinion #2


If you had some very effiicent horn speakers like Klipsh, then maybe.
But with regular cone speakers?

Don't do it.

I need a third opinion here to break the deadlock. BTW, I found out that
the efficiency is 88 db/w/m.


Well, there are opinions and there are opinions.

In the realm of physics, not all opinions are created equal.
The person who made the claims:

"Don't really need to know the efficiency"

"Your tube amp will be cranking at full rated power and
be pushing the speakers at just above conversation level."

Had no technical justification for these rather
extraordinary claims. In addition, he provided numbers
which contradicted his own claims.

If you try it, and you're sane about it, meaning that
if the music starts to sound too loud and distorted,
it IS too loud and distorted, then the risk is VERY small.

Yes, there are opinion. and there are physical facts.

The physical facts are

1. there's a difference of 7 dB between the rated power
of the two amplifiers.

2. At normal to moderately loud listening levels, with
speakers with an 88 dB@1W sensitivity, you're no
where near requiring 10 watts, much less 50 watts.

3. With only 10 watts per speaker, not only is the
smaller amp capable of playing at level far above
"conversational levels", it's capable of playing well
at levels that would make conversation difficult,
e.g., SPL's in the mid 90 dB realm or better in a
typical domestic living situation in a room of 50 m^3
(1300 ft^3) sitting about 3m (9 ft) away from the
speakers.

4. Boston Acoustics stated the speaker required 15
watts minimum power. That's only 1.8 dB more than
10 watts. 1.8 dB is NOT a tremendous difference, and
could pass as less than a just noticeable difference
under uncontrolled, informal, recreational listening.

There are opinions, and there are physical facts. Physical
facts trump opinions pretty much all the time, unless those
opinions are well grounded in the facts.

I gave my opinion and didn't try to disguise it as fact. I'm not here to
act like some pompous know-it-all to show off my incredible knowledge.

You have twisted this argument into an issue of facts verses opinions.
My answer was not to taken as a bonafide, infinitely accurate assesment
of the situation, just as a godd idea of what the original poster may
have in store.

So, yes, I exxagerated the claims of conversation level etc, but it was
to make my point.

Since you are so good at accurately calculating the capabilty of this
scenario, how about taking into account the user's listening tastes in
music and how a 10 watt per channel tube amp with his Boston acoustic
speakers will deal with dynamics? If he likes jazz and a 40 Hz bass note
hits at full scale, how will it sound on this system? If he likes
classical and the orchestra ends a piece with a final crescendo, will it
sound satisfying? I think not. These examples I beleive will sound weak
and leave the listener wanting on that 10 watt system.

My bet is also that, if you make these calculations and prove that I may
be right, you will keep quiet and post nothing. If I am wrong, you
will post the a response and continue to make all effort to dis-credit me.

What say you?

CD

[Arny Krueger]

"Peter Wieck" wrote in message
...

On Oct 26, 5:27 pm, "Arny Krueger" wrote:

a) Tube amps clip softly.

False claim. Tube amps plenty sharply if they have enough loop feedback
to
be usable for hi fi purposes.

Some PR mileage has been made out of the fact that tubed amps often have
relatively poorly-regulated power supplies, but once higher-voltage
transistors became more available, lots of SS amps with relatively
poorly-regulated power supplies have been built and sold.- Hide quoted
text -

Tube amps, at least those that use output tranformers do not pass DC
to the speakers.

Neither do typical solid state amps. They are not DC-coupled. There at least
one LF pole. So here we have yet another straw man argument.

Hence, they clip - SOFTLY - as compared to many/most
vintage SS amps of the same/slightly greater output capacity.

DC coupling has nothing to do with soft clipping. That makes yet another
straw man argument.

And, hence, they are less likely to damage speakers (or themselves) if
driven to clipping, assuming reasonable build quality in either case.

The issue was soft clipping, not probability of damage. That makes three
straw man arguments in just this one post. Or, is it a red herring? BTW, in
the days of tubes we managed to destroy speakers quite nicely with tubed
amps. Most speakers are destroyed by the application of excess power, and it
matters not where that excess power comes from tube amp or SS amp, because
regardless of source it is all turned into heat by the resistance of the
voice coil & etc.

That would be the point. Headroom is a far more critical issue when
discussing SS amps than it is with tube amps due to their very
different natures. A 50-watt solid-state amp, therefore, is not
substantially more powerful in any meaningful way than a 20-watt Tube
amp.

This would appear to be a baseless assertion, given that the assertions that
preceeded it are easy to disregard on the grounds that they are straw man
arguments.

[[email protected]]

On Oct 27, 10:47 am, Sonnova wrote:
I was taught in engineering school that tubes produced
mostly even-order harmonic distortion when they clipped
and that the clipped waveform didn't look as much like a
square wave as does a transistor which, when fully "on"
acts like a switch and produces lots of odd-order distortion.

False.

The issue of whether a circuit produces even-order or
odd-order distortion has far less to do with whether
it's implemented with tubes or solid state than it
has to do whether the non-linearities are symmetrical
or non-symmetrical.

Take the case of a push-pull tube amplifier vs a push-
pull solid state amplifier: at clipping (assuming the
clipping is symmetrical about 0), BOTH implementations
produce essentially odd-order only, simply because the
symmetry is odd-order. Take two single-ended amplifiers,
or two with clip unsymmetrically, one solid sate, the other
tubed, and they BOTH produce even-order distortion.

Single-ended low-global feedback tube amplifiers DO
produce a LOT of even order distortion, but so do single-
ended, low feedback solid state amplifiers as yet another
example.

[Sonnova]

On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

"Peter Wieck" wrote in message
...

On Oct 26, 5:27 pm, "Arny Krueger" wrote:

a) Tube amps clip softly.

False claim. Tube amps plenty sharply if they have enough loop feedback
to
be usable for hi fi purposes.

Some PR mileage has been made out of the fact that tubed amps often have
relatively poorly-regulated power supplies, but once higher-voltage
transistors became more available, lots of SS amps with relatively
poorly-regulated power supplies have been built and sold.- Hide quoted
text -

Tube amps, at least those that use output tranformers do not pass DC
to the speakers.

Neither do typical solid state amps. They are not DC-coupled. There at least
one LF pole. So here we have yet another straw man argument.

Hence, they clip - SOFTLY - as compared to many/most
vintage SS amps of the same/slightly greater output capacity.

DC coupling has nothing to do with soft clipping. That makes yet another
straw man argument.

And, hence, they are less likely to damage speakers (or themselves) if
driven to clipping, assuming reasonable build quality in either case.

The issue was soft clipping, not probability of damage. That makes three
straw man arguments in just this one post. Or, is it a red herring? BTW, in
the days of tubes we managed to destroy speakers quite nicely with tubed
amps. Most speakers are destroyed by the application of excess power, and it
matters not where that excess power comes from tube amp or SS amp, because
regardless of source it is all turned into heat by the resistance of the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are damaged
by too little power as by too much. Here's the scenario: Small amp is driven
into constant clipping by listener's desire for louder rock-n-roll. Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil. Coil
heats-up, burns out.

[Arny Krueger]

"Sonnova" wrote in message
...

I was taught in engineering school that tubes produced mostly even-order
harmonic distortion

No even order distortion if those tubes are in a push-pull configuration,
which has been the standard for power amps since about 1933.

when they clipped and that the clipped waveform didn't
look as much like a square wave

Tubes cut off and saturate just like transistors. The edges of a clipped
signal are sharpened by inverse feedback, which has been the standard since
about 1938.

as does a transistor which, when fully "on"
acts like a switch

Tubes also act like switches. Just drive them hard enough.

and produces lots of odd-order distortion.

Odd order distortion is a natural consequence of clipping of both the
negative and positve halves of the signal.

Perhaps, since
even-order harmonic distortion is easier on the ear and more consonant
with
music, that this is the source of the term "soft-clipping".

Repeat: push-pull amplifiers can't produce even order distortion whether
they are tubed or SS, because of the rule of half wave symmetry,

Of course, driven
hard into clipping, no amplifying device is listenable, no matter what
type
of distortion is being produced but momentary peaks are another matter.

Actually, if those momentary peaks are really short, they pretty well escape
the ear's notice if we're amplifying music.

Please name the engineering school that taught you all of these strange
things.

[Arny Krueger]

wrote in message
...

Single-ended low-global feedback tube amplifiers DO
produce a LOT of even order distortion, but so do single-
ended, low feedback solid state amplifiers as yet another
example.

http://en.wikipedia.org/wiki/Tube_sound#Soft_clipping

"However the monotonically reducing purely harmonic distortion spectrum of
the simple SE gain stage is not of itself anything to do with using a tube,
and a similar distortion spectrum and sound can be obtained using the same
circuit topology with , for example, a MOSFET."

" However, soft clipping is not exclusive to tubes"

Case in point being Nelson Pass' Zen series of SS power amplifiers.

[Sonnova]

On Sat, 27 Oct 2007 20:08:26 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...

I was taught in engineering school that tubes produced mostly even-order
harmonic distortion

No even order distortion if those tubes are in a push-pull configuration,
which has been the standard for power amps since about 1933.

when they clipped and that the clipped waveform didn't
look as much like a square wave

Tubes cut off and saturate just like transistors. The edges of a clipped
signal are sharpened by inverse feedback, which has been the standard since
about 1938.

as does a transistor which, when fully "on"
acts like a switch

Tubes also act like switches. Just drive them hard enough.

and produces lots of odd-order distortion.

Odd order distortion is a natural consequence of clipping of both the
negative and positve halves of the signal.

Perhaps, since
even-order harmonic distortion is easier on the ear and more consonant
with
music, that this is the source of the term "soft-clipping".

Repeat: push-pull amplifiers can't produce even order distortion whether
they are tubed or SS, because of the rule of half wave symmetry,

Of course, driven
hard into clipping, no amplifying device is listenable, no matter what
type
of distortion is being produced but momentary peaks are another matter.

Actually, if those momentary peaks are really short, they pretty well escape
the ear's notice if we're amplifying music.

Please name the engineering school that taught you all of these strange
things.


San Jose State University.

[Arny Krueger]

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with tubed
amps. Most speakers are destroyed by the application of excess power, and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil. Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound muddy.

[Sonnova]

On Sun, 28 Oct 2007 20:07:40 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with tubed
amps. Most speakers are destroyed by the application of excess power, and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil. Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound muddy.


Seen too many burned-out tweeters associated with too-small amps to buy that.
Sorry.

[Randy Yates]

"Arny Krueger" writes:
[...]
The fallacy here is the idea that clipped musical signals always have more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals.

Tell me where the following thought experiment (which I believe should
model reality) falls apart then: 1) A signal is composed of a set of
sinusoids. 2) Clipping any one sinuoid will produce other sinusoids that
are at integer multiples of the original (i.e., that are "high
frequency") 3) These other sinusoids will be "extra" if they did not
exist before.

It is possible that the new sinusoids will add destructively with
existing sinusoids, but I claim such occurrences are "pathological
cases."
--
% Randy Yates % "Midnight, on the water...
%% Fuquay-Varina, NC % I saw... the ocean's daughter."
%%% 919-577-9882 % 'Can't Get It Out Of My Head'
%%%% % *El Dorado*, Electric Light Orchestra
http://www.digitalsignallabs.com

[Arny Krueger]

"Sonnova" wrote in message
...
On Sun, 28 Oct 2007 20:07:40 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with
tubed
amps. Most speakers are destroyed by the application of excess power,
and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of
the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil. Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In
fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound muddy.

Seen too many burned-out tweeters associated with too-small amps to buy
that.

How does a tweeter show that it was burned out by too small of an amplifier?

[Arny Krueger]

"Randy Yates" wrote in message
...
"Arny Krueger" writes:

[...]
The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals.

Tell me where the following thought experiment (which I believe should
model reality) falls apart then: 1) A signal is composed of a set of
sinusoids. 2) Clipping any one sinuoid will produce other sinusoids that
are at integer multiples of the original (i.e., that are "high
frequency") 3) These other sinusoids will be "extra" if they did not
exist before.

Your thought experiment is true as far as it goes, but it ignores a major
effect of clipping on signals related to music. One important aspect of
signal analysis is to actually observe real world signals. Since I do a lot
of digital editing of music, I look at music waveforms for hours and hours.
Every once in a while I get levels wrong while doing live recording, and
then I get to look at clipped music. I make frequent use of spectral
analysis while I'm editing, particularly if I'm editing noisy recordings, or
using equalization.

The spectral content of music very often decreases with frequency. Some
very harsh rock music actually has uniform energy per octave or frequency,
but most music has an energy distribution that resembles pink or red noise.

Clipping is most often stimulated by the low frequency components of the
signal because the low frequency components of music tend to have the
highest amplitude of any of its component signals.

When you clip a signal you create segments where the signal is a essentually
a DC pulse whose duration is related to the frequency of the signal that
stimulated the clipping. This replaces the high frequency components of the
music with a straight line. They are effectively erased.

IOW, any treble notes that would normally ride on top of a bass note are
replaced with straight lines when the bass note is clipped. The treble notes
are eliminated by the clipping. High frequency energy is reduced.

Higher frequency components that once existed in the clipped signal segment
are effectively erased by the clipping.

Square waves have a spectral content that is similar to red noise. If the
unclipped music had a spectral content that was similar to white or pink
noise, then its high frequency content will have been reduced. If the
unclipped music had a spectral content that was similar to red noise, then
its high frequency content is essentially unchanged.

Rarely does clipping ever actually increase the high frequency content of
the music.

[Sonnova]

On Tue, 30 Oct 2007 17:22:04 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sun, 28 Oct 2007 20:07:40 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with
tubed
amps. Most speakers are destroyed by the application of excess power,
and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of
the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil. Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In
fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound muddy.

Seen too many burned-out tweeters associated with too-small amps to buy
that.

How does a tweeter show that it was burned out by too small of an amplifier?


Simple. Hook a bookshelf speaker system rated at "200 watts" to a 25
Watt/channel Wiliamson amplifier and notice that it keeps blowing tweeters
(and is not oscillating).

[Andrew Haley]

Randy Yates writes:
"Arny Krueger" writes:
[...]
The fallacy here is the idea that clipped musical signals always have more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals.

Tell me where the following thought experiment (which I believe should
model reality) falls apart then: 1) A signal is composed of a set of
sinusoids. 2) Clipping any one sinuoid will produce other sinusoids that
are at integer multiples of the original (i.e., that are "high
frequency") 3) These other sinusoids will be "extra" if they did not
exist before.

Yes, but experiment shows that even when an amplifier is grossly
overdriven, although the clipping is producing extra harmonics they
never approach the levels of the amplified high frequency source
signal.

There's a famous experiment at RaneNote 128:

http://www.audiovisualdevices.com.au...ne/note128.pdf

Andrew.

[Randy Yates]

"Arny Krueger" writes:
[...]

Thanks Arny. I think the problem with my gedanken is that I assume the
system is linear. An amplifier driven into clipping is not a linear
system!

That is essentially why, e.g., if you have a small high frequency sine
wave riding on a low frequency sine wave and you clip the low-frequency
sine wave, you lose the high frequency sine wave, which was higher in
amplitude than the clipped waveforms harmonics.
--
% Randy Yates % "With time with what you've learned,
%% Fuquay-Varina, NC % they'll kiss the ground you walk
%%% 919-577-9882 % upon."
%%%% % '21st Century Man', *Time*, ELO
http://www.digitalsignallabs.com

[Arny Krueger]

"Sonnova" wrote in message
...
On Tue, 30 Oct 2007 17:22:04 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sun, 28 Oct 2007 20:07:40 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with
tubed
amps. Most speakers are destroyed by the application of excess power,
and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of
the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil.
Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In
fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound
muddy.

Seen too many burned-out tweeters associated with too-small amps to buy
that.

How does a tweeter show that it was burned out by too small of an
amplifier?


Simple. Hook a bookshelf speaker system rated at "200 watts" to a 25
Watt/channel Wiliamson amplifier and notice that it keeps blowing tweeters
(and is not oscillating).

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts.

[Arny Krueger]

"Andrew Haley" wrote in message
...
Randy Yates writes:
"Arny Krueger" writes:
[...]
The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals.

Tell me where the following thought experiment (which I believe should
model reality) falls apart then: 1) A signal is composed of a set of
sinusoids. 2) Clipping any one sinuoid will produce other sinusoids that
are at integer multiples of the original (i.e., that are "high
frequency") 3) These other sinusoids will be "extra" if they did not
exist before.

Yes, but experiment shows that even when an amplifier is grossly
overdriven, although the clipping is producing extra harmonics they
never approach the levels of the amplified high frequency source
signal.

There's a famous experiment at RaneNote 128:

http://www.audiovisualdevices.com.au...ne/note128.pdf

You're no doubt thinking of figures 3 and 4.

This shows the effect that I described in my reply to Yates's thought
experiment - where clipping actually removes high frequency energy from the
musical signal.

I've replicated this process with numerous examples of real world music and
obtained similar results. It's a pretty easy experiment to do with 16 bit
files in Audition/CEP . You just do a spectrum analysis of some music,
over-amplify it to clip it, (20 dB of gain will usually suffice) and then
repeat the spectrum analysis. Audition has a nifty curve hold feature that
facilitates the comparison. The amplification will increase the overall
height of the second curve, but on the average it will be roughly parallel
to the held curve for the unclipped signal. Or, you can start out with a
mono signal, clone it into 2 channels, and clip just one of the channels.

Often, there is no significant change in the energy distribution of the
music at all when it is clipped.

OTOH, there will probably be a general increase in broadband energy levels
in the clipped music, because clipping also provides a sort of dynamics
compression effect. Clipped signals tend to optimize the broadband average
power levels that a given amplifier can deliver. If you want to get the most
power out of an amp that it will deliver, clip the music pretty heavily.

So what it comes down to is that by clipping music, its average broadband
power is usually increased, while the proportion of energy at high
frequencies will remain about the same or may even decrease.

If there is a decrease in energy in the upper midrange where the ear is most
sensitive, subjective loudness for a given power level might even decrease.

Of course, when you increase power at all frequencies, you also increase
power at high frequencies.

It's the increased average power at all frequencies that burns out drivers,
not some highly technical shifting of energy from low frequencies to high
frequencies.

That's the point of Rane note 128, which has AFAIK never been technically
rebutted over the approximately 15 years that it has been published.

[Peter Wieck]

On Nov 1, 6:08 pm, "Arny Krueger" wrote:

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts

Which means exactly nothing in that same "real world". Send 10 watts
of power as a sine-wave to any speaker for any sustained period of
time and I suspect that you would be able to blow your driver-of-
choice in a very few minutes, if even that. With VDF (very ****** few)
exceptions.

However and at the same time, take the average stock-but-working first-
run Dynaco ST-120 driven to clipping 60% of the time and it will be a
close-run thing whether the speakers or the amp fails first. At that
point, sure the tweeter is likely cheaper than the woofer but does it
really matter which driver it is?

Peter Wieck
Wyncote, PA

[Sonnova]

On Thu, 1 Nov 2007 16:08:10 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Tue, 30 Oct 2007 17:22:04 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sun, 28 Oct 2007 20:07:40 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Sat, 27 Oct 2007 07:51:03 -0700, Arny Krueger wrote
(in article ):

BTW, in
the days of tubes we managed to destroy speakers quite nicely with
tubed
amps. Most speakers are destroyed by the application of excess power,
and
it
matters not where that excess power comes from tube amp or SS amp,
because
regardless of source it is all turned into heat by the resistance of
the
voice coil & etc.

Actually, I think you'll find that just as many speaker systems are
damaged
by too little power as by too much. Here's the scenario: Small amp is
driven
into constant clipping by listener's desire for louder rock-n-roll.
Clipped
high frequencies exceed the duty cycle of the tweeter's voice coil.
Coil
heats-up, burns out.

The fallacy here is the idea that clipped musical signals always have
more
high frequencies than unclipped musical signals. Clipping does not
necessarily increase the high frequency content of musical signals. In
fact,
in some cases, clipping decreases the high frequency content of musical
signals. That's one reason why clipping tends to make music sound
muddy.

Seen too many burned-out tweeters associated with too-small amps to buy
that.

How does a tweeter show that it was burned out by too small of an
amplifier?


Simple. Hook a bookshelf speaker system rated at "200 watts" to a 25
Watt/channel Wiliamson amplifier and notice that it keeps blowing tweeters
(and is not oscillating).

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts.


Most tweeters are like that, that's (one reason) why they have at least a
capacitor in series with them and sometimes a fuse. The fuse doesn't usually
blow in these cases because the cooked tweeter is not due to excess current,
but rather because it's voice-coil duty cycle has been continuously exceeded
by a clipping amplifier.

[Sonnova]

On Thu, 1 Nov 2007 16:22:21 -0700, Peter Wieck wrote
(in article ):

On Nov 1, 6:08 pm, "Arny Krueger" wrote:

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts

Which means exactly nothing in that same "real world". Send 10 watts
of power as a sine-wave to any speaker for any sustained period of
time and I suspect that you would be able to blow your driver-of-
choice in a very few minutes, if even that. With VDF (very ****** few)
exceptions.

However and at the same time, take the average stock-but-working first-
run Dynaco ST-120 driven to clipping 60% of the time and it will be a
close-run thing whether the speakers or the amp fails first. At that
point, sure the tweeter is likely cheaper than the woofer but does it
really matter which driver it is?

Peter Wieck
Wyncote, PA

I had an early ST-120. What a terrible device! Dynaco eventually fixed it,
but it was, originally, one of the most fragile and hard sounding amplifiers
I've ever seen (I must have had to replace the output and driver transistors
at least 5 times in as many years). With it's matching preamp, the PAT-5, its
hard to tell which was worse the amp or the preamp. Dynaco called it the
"transistor sound", I called it distortion. I replaced those horrid
components with an Audio Research SP-3 and pair of D-40's and never looked
back! I still prefer tubes to this day and see no reason to change (my subs
have SS amps in them though). My current main amps, a pair of VTL 140
monoblocs, though 15 years old, are so good, that I'll probably keep 'em
forever. I have purchased, from an old ham operator, a "lifetime supply" of
WWII vintage NOS JAN 807s (with the brown bakelite tube bases). 807s are used
by these amps (6 each) as output tubes and I have found that the JAN 807s
sound better (and are certainly better made) than the Chinese versions that
VTL sells as replacements - and which came in the amps when new. They drive a
pair of Martin-Logan Vantage loudspeakers with built-in powered subwoofers.
Glorious sound!

[http://www.wenterprisesnorthwest.com]

To get right to the point, unless you have a large room, 10-20 wpc of tube
amplification is sufficient to drive your T-830s (88dB @ 2.83V/1m) to loud
volumes.

There is a review (see review #4) that specifically talks about your
speakers with a 20 wpc integrated NAD solid-state amp...

http://www.audioreview.com/cat/ampli...7_1593crx.aspx

On top of that, if you prefer the smooth sound of tubes, you will like that
set-up better even if it may not get quite as loud as your H/K solid-state.
The only downfall I could see with switching to this amp with less power
could be deep bass power since it uses most of the power to reproduce
especially at higher volumes.

Happy listening!

"Dave" wrote in message
...
A acquaintance of mine is wanting to upgrade his amplifier, an EL34-based
tube amp wired in ultralinear configuration. I'm considering buying it,
but
am unsure if this amp's 20 wpc is enough to drive my Boston Acoustics
T-830
speakers. Right now they're driven with a 50 wpc Harmon Kardon vintage
solid state amp, and that setup can be played LOUD. If I purchased the
tube
amplifier, I'd likely want to rewire it in triode configuration which
would
halve my power output to around 10 wpc.

Does anyone know the efficiency specs for the T-830 speakers, or any of
the
T-series BA loudspeakers from the early 90's? I have the original cartons
but none of the literature that came with the speakers.

Thanks

Dave S.

[Arny Krueger]

"Peter Wieck" wrote in message
...
On Nov 1, 6:08 pm, "Arny Krueger" wrote:

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts

Which means exactly nothing in that same "real world". Send 10 watts
of power as a sine-wave to any speaker for any sustained period of
time and I suspect that you would be able to blow your driver-of-
choice in a very few minutes, if even that. With VDF (very ****** few)
exceptions.

A lot of modern tweeters are actually very robust. Robust = fewer warranty
claims. Innovations like Ferrofluid, high-temperature insulation and glues
have made a big difference. Adhesives and varnishes may have improved the
most - epoxies that retain their strength at 600 degrees or more are now
just a matter of ordering up the right stuff. Old-time varnishes and glues
started loosing strength on a hot sunny day!

If *any speaker* includes modern high-performance compression tweeter
drivers, then continuous power handling in the 40 to 100 watt range is
possible. Of course, you get to pay for performance like that! ;-)

If *any speaker* includes midranges and woofers, then the sky is literally
the limit. Some modern woofers will sit and stroke with 100s of watts at
20-30 Hz or higher, all day long.

However and at the same time, take the average stock-but-working first-
run Dynaco ST-120 driven to clipping 60% of the time and it will be a
close-run thing whether the speakers or the amp fails first.

My Dyna ST-120 has proven to be more robust than that. While everything
including output transistors appars to be original, I don't know where it
sat in the ST-120 production run. I suspect that it is a factory build.
During the life of the ST-120 as a product, SS technology made large
advances in power transistor speed, linearity, costs and robustness.

[Sonnova]

On Fri, 2 Nov 2007 15:34:13 -0700, Arny Krueger wrote
(in article ):

"Peter Wieck" wrote in message
...
On Nov 1, 6:08 pm, "Arny Krueger" wrote:

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts

Which means exactly nothing in that same "real world". Send 10 watts
of power as a sine-wave to any speaker for any sustained period of
time and I suspect that you would be able to blow your driver-of-
choice in a very few minutes, if even that. With VDF (very ****** few)
exceptions.

A lot of modern tweeters are actually very robust. Robust = fewer warranty
claims. Innovations like Ferrofluid, high-temperature insulation and glues
have made a big difference. Adhesives and varnishes may have improved the
most - epoxies that retain their strength at 600 degrees or more are now
just a matter of ordering up the right stuff. Old-time varnishes and glues
started loosing strength on a hot sunny day!

If *any speaker* includes modern high-performance compression tweeter
drivers, then continuous power handling in the 40 to 100 watt range is
possible. Of course, you get to pay for performance like that! ;-)

If *any speaker* includes midranges and woofers, then the sky is literally
the limit. Some modern woofers will sit and stroke with 100s of watts at
20-30 Hz or higher, all day long.

However and at the same time, take the average stock-but-working first-
run Dynaco ST-120 driven to clipping 60% of the time and it will be a
close-run thing whether the speakers or the amp fails first.

My Dyna ST-120 has proven to be more robust than that. While everything
including output transistors appars to be original, I don't know where it
sat in the ST-120 production run. I suspect that it is a factory build.
During the life of the ST-120 as a product, SS technology made large
advances in power transistor speed, linearity, costs and robustness.


I know that in the early seventies, Dynaco changed the ST-120 considerably.
They replaced the output transistors and the drivers with devices that were
more robust (the originals were working at their ragged edge, so much so that
they had to be hand-selected for the task by Dyna and had special part
numbers. If a repair tech bought off-the-shelf transistors of the proper
type, they generally blew fairly soon after the amp was put back in service.
Also, if one driver or output transistor went on the ST-120, it took the
other three of those four transistors with it - almost every time! I also
know that the later ST-120s were biased harder into class AB than the earlier
amps had been. They had to do something. As I said, my Gen 1 ST-120/PAT-5
sounded simply dreadful!

[Arny Krueger]

"Sonnova" wrote in message
...

On Thu, 1 Nov 2007 16:08:10 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...

Simple. Hook a bookshelf speaker system rated at "200 watts" to a 25
Watt/channel Wiliamson amplifier and notice that it keeps blowing
tweeters
(and is not oscillating).

The bookshelf tweeter probably has a real-world (not spec sheet) power
capacity of under 10 watts.

Most tweeters are like that,

Not these days, not in quality speakers. Please see my other recent post on
the matter.

that's (one reason) why they have at least a
capacitor in series with them

The capacitor is a part of a filter that routes lows to the woofer and highs
to the tweeters in the interest of smooth response.

and sometimes a fuse.

Or other part like a light bulb that protects the tweeter from abuse.

The fuse doesn't usually
blow in these cases because the cooked tweeter is not due to excess
current,
but rather because it's voice-coil duty cycle has been continuously
exceeded
by a clipping amplifier.

Excess current and exceeding the duty cycle are two ways of describing the
identical same process.