Jim Lesurf wrote:


FWIW I use +200Wpc amps with speakers like ESL63's and before that ESL57's.
Been doing this for around 20 years, now. Generally I only play at quite
low mean power levels. The results sound fine to me even though I never go
anywhere near 200 Wpc. Having an amp that can drive difficult loads means
the amp is finding this task easy, even when I once used parallel pairs of
ESL57's for a while. :-)



Any idea what's the best way to stack ESL57's?
Should they be coplanar or focussed at the listener's head?

--
Roger.

On Sun, 04 Jan 2004 17:32:20 +0000, Stewart Pinkerton wrote:

On Sun, 04 Jan 2004 12:31:32 GMT, "Form@C" wrote:

On Sun, 04 Jan 2004 11:09:48 +0000, Ian Molton wrote:

On Sun, 04 Jan 2004 08:27:50 GMT
"Form@C" wrote:

It's no use turning down a
100-200W monster amp because you arn't running them at their best.

what?

Sorry, it wasn't very clear was it? What I meant to say is that "high"
power amps are usually designed and set up to produce their best results
an appreciable way up to their rated power output.

No, they aren't.


Oh yes they are... You will almost always see distortion levels etc
quoted for a particular power output level. That power level is often
quoted at either full rated power or at least somewhere well above 5W
or so - very loud in most domestic rooms. (You may also notice that
manufacturers don't often quote exactly what the load was theat the amp
was driving when the measurements were taken. Some amps measure well into
resistive loads only.) Only the better amps will quote specifications at
low power levels.


They won't perform at
their best at a very low percentage of that level.

Absolute garbage!


Maybe, but its a fact! :-)


I *know* that many are
*supposed* to be class A at low levels, but they are generally not
designed as class A amps, and are not really running as such (the output
stage quiescent current is usually too low to allow correct class A
operation on a class AB amp).

Excuse me? A class AB amp *always* runs in class A up to some
predetermined level, which may be less than 1 watt or as many as 50 watts.
I use an amplifier which runs in true class A up to 50 watts into 8 ohms,
and in class AB above that level, up to its rated maximum of 400 watts
into 1 ohm. It sounds just fine at very low levels.


You are correct, but that predetermined level can be very vague in many
cases. There is no way for the user to tell when the amp changes out of
class A without getting a scope out, so manufacturers often "bend" the
quoted output to make the specs look good! An amp can be classed as AB if
it produces the first 50mW as class A of course... At 400W you are not
running an amp that I would class as average domestic. It will (should) be
running bloody hot too! (1) The majority of AB amps out there run the
no-load current at below optimum for usable class A use and don't apply
sufficient drive to push the output devices fully. It keeps the size &
cost of the PSU down and reduces the heat, making the equipment more
acceptable to the domestic market.


Thus it is a mistake to judge low-power
audio systems by simply turning down the volume on existing "high" power
equipment & changing the speakers for more sensitive ones.

Depends on the amp......................

Oh yeah! There are a lot of really good amps, but you won't often find
them available through your local outlets. Even so-called "hi-fi" shops
tend not to carry the good stuff as stock - it costs far too much.



(1) A note for thos who may not know what we are on about... Stewart will
know this already! Class A is always less than 50% efficient, typically
46%, so you should be producing well over 100W of heat, with no signal
present, to be capable of class A at 50W - not including valve heaters of
course! In class AB amplifiers one of the output devices is cut-off on
alternate halves of the waveform, giving a power saving. Class B
amplifiers always cut off one of the devices at some point. This is the
primary cause of the "transistor sound" - a lot of odd-harmonic
distortion.

--
Mick
http://www.nascom.info for Nascom & Gemini information
Also at http://www.mixtel.co.uk where the collection started.
Currently deserting M$ for linux... :-)

On Sun, 04 Jan 2004 17:32:20 +0000, Stewart Pinkerton wrote:

On Sun, 04 Jan 2004 12:31:32 GMT, "Form@C" wrote:

On Sun, 04 Jan 2004 11:09:48 +0000, Ian Molton wrote:

On Sun, 04 Jan 2004 08:27:50 GMT
"Form@C" wrote:

It's no use turning down a
100-200W monster amp because you arn't running them at their best.

what?

Sorry, it wasn't very clear was it? What I meant to say is that "high"
power amps are usually designed and set up to produce their best results
an appreciable way up to their rated power output.

No, they aren't.


Oh yes they are... You will almost always see distortion levels etc
quoted for a particular power output level. That power level is often
quoted at either full rated power or at least somewhere well above 5W
or so - very loud in most domestic rooms. (You may also notice that
manufacturers don't often quote exactly what the load was theat the amp
was driving when the measurements were taken. Some amps measure well into
resistive loads only.) Only the better amps will quote specifications at
low power levels.


They won't perform at
their best at a very low percentage of that level.

Absolute garbage!


Maybe, but its a fact! :-)


I *know* that many are
*supposed* to be class A at low levels, but they are generally not
designed as class A amps, and are not really running as such (the output
stage quiescent current is usually too low to allow correct class A
operation on a class AB amp).

Excuse me? A class AB amp *always* runs in class A up to some
predetermined level, which may be less than 1 watt or as many as 50 watts.
I use an amplifier which runs in true class A up to 50 watts into 8 ohms,
and in class AB above that level, up to its rated maximum of 400 watts
into 1 ohm. It sounds just fine at very low levels.


You are correct, but that predetermined level can be very vague in many
cases. There is no way for the user to tell when the amp changes out of
class A without getting a scope out, so manufacturers often "bend" the
quoted output to make the specs look good! An amp can be classed as AB if
it produces the first 50mW as class A of course... At 400W you are not
running an amp that I would class as average domestic. It will (should) be
running bloody hot too! (1) The majority of AB amps out there run the
no-load current at below optimum for usable class A use and don't apply
sufficient drive to push the output devices fully. It keeps the size &
cost of the PSU down and reduces the heat, making the equipment more
acceptable to the domestic market.


Thus it is a mistake to judge low-power
audio systems by simply turning down the volume on existing "high" power
equipment & changing the speakers for more sensitive ones.

Depends on the amp......................

Oh yeah! There are a lot of really good amps, but you won't often find
them available through your local outlets. Even so-called "hi-fi" shops
tend not to carry the good stuff as stock - it costs far too much.



(1) A note for thos who may not know what we are on about... Stewart will
know this already! Class A is always less than 50% efficient, typically
46%, so you should be producing well over 100W of heat, with no signal
present, to be capable of class A at 50W - not including valve heaters of
course! In class AB amplifiers one of the output devices is cut-off on
alternate halves of the waveform, giving a power saving. Class B
amplifiers always cut off one of the devices at some point. This is the
primary cause of the "transistor sound" - a lot of odd-harmonic
distortion.

--
Mick
http://www.nascom.info for Nascom & Gemini information
Also at http://www.mixtel.co.uk where the collection started.
Currently deserting M$ for linux... :-)

On Sun, 04 Jan 2004 17:17:32 +0000, Stewart Pinkerton wrote:

On Sun, 04 Jan 2004 08:27:50 GMT, "Form@C" wrote:

Have a look at any mid-price, fairly sensitive, half-decent speakers &
couple them with a little pure class A amp. Valves are nice...

No, they aren't - if you want *accurate* reproduction.


Maybe not, but do you want *accurate* reproduction or an amp that is good
to listen to? I, personally, would much rather listen to 2nd harmonic
distortion rather than 3rd!


(you
could even build your own! I've just built a MOSFET headphone amp
that would scale up quite easily.

And what have MOSFETs to do with valves?


Nothing at all. They are far easier for those who want to build small
amplifiers though. You need a decent mains tranny to source the current
for class A, but ther is no OPT to bother with. You can still burn your
hand on the heatsink instead of the valve though. :-)


I bet it only cost me £30 or so if you include the bits from the
"scrap box".) You may be surprised at how little power you really
need; most people can get away with about 3W to 5W per channel under
real-life surroundings.

During the "average" situation maybe. Lets say a symphony orchestra
burbling away steadily. Then comes the climax. One or two huge peak
demands. Maybe several hundred watts. It is the 5W valve amp's
inability to cope with peak requirements without distortion which is
it's weakness.

But in this case the OP wants a more neighbour-friendly system... :-)
We arn't allowed to aim at full concert-hall volume here! In any case,
how close would you be sitting to the orchestra to reach the dB level
for "several hundred watts" *at your seat*?

That's typically 100-105dB, so it depends on your room and your
speakers.


So, given fairly inefficient speakers of about 60dB/W you *need* less than
2W to reach that SPL. Double it to add some headroom. Say 5W. Even small
bookshelf speakers can be about 90dB/W, so this is loud.


Remember to take the sqrt of the
power each time you double the distance back. You tend to sit far closer
to speakers than you do to the orchestra in a concert hall.

And the square law no longer applies once you are half way down the
room..................


No, it tends to get worse because of sound absorbtion (although there is
some reflected stuff too), so the SPL at your listening position is
probably less...



Try it. Use a
*good* low power amp and sensitive speakers. It's no use turning down a
100-200W monster amp because you arn't running them at their best. They
will be set up, probably anyway, to sound best at about 25-50% of full
output.

Absolute garbage!

Maybe for your amp... :-)



Remember that to double the volume you have
to square the power,

No, you need ten times the power.


Sorry, my mistake! I got carried away with sqrts....

so 250W is about twice the volume of a 25W amp, which in turn is
about twice the volume of a 5W amp! The more sensitive your speakers
are, the better.

Yes, but you still need good amplifier power to ensure adequate headroom
for peaks.


If you put an orchestral peak at, say, 200dB (way too high, I know - it
would deafen the conductor.), the SPL at your listening position at, say,
197dB (lost half of it) and your speaker efficiency at, say, 60dB/w
(fairly poor sealed boxes) then you need 4W to hit the peak volume. A 5W
amp is overkill. :-)

--
Mick
http://www.nascom.info for Nascom & Gemini information
Also at http://www.mixtel.co.uk where the collection started.
Currently deserting M$ for linux... :-)

On Sun, 04 Jan 2004 17:17:32 +0000, Stewart Pinkerton wrote:

On Sun, 04 Jan 2004 08:27:50 GMT, "Form@C" wrote:

Have a look at any mid-price, fairly sensitive, half-decent speakers &
couple them with a little pure class A amp. Valves are nice...

No, they aren't - if you want *accurate* reproduction.


Maybe not, but do you want *accurate* reproduction or an amp that is good
to listen to? I, personally, would much rather listen to 2nd harmonic
distortion rather than 3rd!


(you
could even build your own! I've just built a MOSFET headphone amp
that would scale up quite easily.

And what have MOSFETs to do with valves?


Nothing at all. They are far easier for those who want to build small
amplifiers though. You need a decent mains tranny to source the current
for class A, but ther is no OPT to bother with. You can still burn your
hand on the heatsink instead of the valve though. :-)


I bet it only cost me £30 or so if you include the bits from the
"scrap box".) You may be surprised at how little power you really
need; most people can get away with about 3W to 5W per channel under
real-life surroundings.

During the "average" situation maybe. Lets say a symphony orchestra
burbling away steadily. Then comes the climax. One or two huge peak
demands. Maybe several hundred watts. It is the 5W valve amp's
inability to cope with peak requirements without distortion which is
it's weakness.

But in this case the OP wants a more neighbour-friendly system... :-)
We arn't allowed to aim at full concert-hall volume here! In any case,
how close would you be sitting to the orchestra to reach the dB level
for "several hundred watts" *at your seat*?

That's typically 100-105dB, so it depends on your room and your
speakers.


So, given fairly inefficient speakers of about 60dB/W you *need* less than
2W to reach that SPL. Double it to add some headroom. Say 5W. Even small
bookshelf speakers can be about 90dB/W, so this is loud.


Remember to take the sqrt of the
power each time you double the distance back. You tend to sit far closer
to speakers than you do to the orchestra in a concert hall.

And the square law no longer applies once you are half way down the
room..................


No, it tends to get worse because of sound absorbtion (although there is
some reflected stuff too), so the SPL at your listening position is
probably less...



Try it. Use a
*good* low power amp and sensitive speakers. It's no use turning down a
100-200W monster amp because you arn't running them at their best. They
will be set up, probably anyway, to sound best at about 25-50% of full
output.

Absolute garbage!

Maybe for your amp... :-)



Remember that to double the volume you have
to square the power,

No, you need ten times the power.


Sorry, my mistake! I got carried away with sqrts....

so 250W is about twice the volume of a 25W amp, which in turn is
about twice the volume of a 5W amp! The more sensitive your speakers
are, the better.

Yes, but you still need good amplifier power to ensure adequate headroom
for peaks.


If you put an orchestral peak at, say, 200dB (way too high, I know - it
would deafen the conductor.), the SPL at your listening position at, say,
197dB (lost half of it) and your speaker efficiency at, say, 60dB/w
(fairly poor sealed boxes) then you need 4W to hit the peak volume. A 5W
amp is overkill. :-)

--
Mick
http://www.nascom.info for Nascom & Gemini information
Also at http://www.mixtel.co.uk where the collection started.
Currently deserting M$ for linux... :-)

On Sun, 04 Jan 2004 19:10:08 GMT
"Form@C" wrote:

There is no way for the user to tell when the amp changes out of
class A without getting a scope out

If that is truely the case then the point becomes utterly moot.

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

On Sun, 04 Jan 2004 19:10:08 GMT
"Form@C" wrote:

There is no way for the user to tell when the amp changes out of
class A without getting a scope out

If that is truely the case then the point becomes utterly moot.

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

"Form@C" wrote
If you put an orchestral peak at, say, 200dB (way too high, I know - it
would deafen the conductor.),

If by "dB" you mean the normal decibel used to measuring sound
pressure level (20 * log_10(p / 2.0E-5) ), then it would not
deafen the conductor. It would kill him and pulp his body.

the SPL at your listening position at, say,
197dB (lost half of it) and your speaker efficiency at, say, 60dB/w
(fairly poor sealed boxes) then you need 4W to hit the peak volume. A 5W
amp is overkill. :-)

Erm, no.

You should not use the notation "dB/W" because it doesn't work like
that. Correctly, you should specify speaker sensitivity as "X dB at 1W
input power" (at some frequency). Some manufacturers instead say
"X dB at 2.83V" (being 2*pi) RMS input (at some frequency), which is a
similar thing, because P = V^2 / Z. If R=8 Ohms at the test frequency,
then the power will be 1 watt. But only if Z=8 Ohms at that frequency,
and in general, it won't be.


Martin
--
M.A.Poyser Tel.: 07967 110890
Manchester, U.K. http://www.fleetie.demon.co.uk

"Form@C" wrote
If you put an orchestral peak at, say, 200dB (way too high, I know - it
would deafen the conductor.),

If by "dB" you mean the normal decibel used to measuring sound
pressure level (20 * log_10(p / 2.0E-5) ), then it would not
deafen the conductor. It would kill him and pulp his body.

the SPL at your listening position at, say,
197dB (lost half of it) and your speaker efficiency at, say, 60dB/w
(fairly poor sealed boxes) then you need 4W to hit the peak volume. A 5W
amp is overkill. :-)

Erm, no.

You should not use the notation "dB/W" because it doesn't work like
that. Correctly, you should specify speaker sensitivity as "X dB at 1W
input power" (at some frequency). Some manufacturers instead say
"X dB at 2.83V" (being 2*pi) RMS input (at some frequency), which is a
similar thing, because P = V^2 / Z. If R=8 Ohms at the test frequency,
then the power will be 1 watt. But only if Z=8 Ohms at that frequency,
and in general, it won't be.


Martin
--
M.A.Poyser Tel.: 07967 110890
Manchester, U.K. http://www.fleetie.demon.co.uk

"Stewart Pinkerton" wrote in message
...
On Sun, 4 Jan 2004 00:01:48 +0000 (UTC), "David"
wrote:

No. Highly efficient speakers are generally less accurate than "lowish"
efficiency speakers. They tend to become unlinear at high outputs.

Actually, the exact opposite is true. It's insensitive speakers that
have problems at high SPLs, for reasons which should be obvious.

Erm, remind me!!




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----