Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment, so
how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall impedance
does.

Graham

Mike Rieves wrote:
Scott Dorsey writes:
As for me, I'm still using the Magnapans, which are just frighteningly
close to a 6 ohm resistive load.

Magnepans can deliver great sonic realism in the right room with the right
source material, but they just aren't very good studio monitor speakers. :-)

Strange, I've been mixing on them for about twenty years now and I've been
very happy. I like the distant presentation a lot; since I tend to like a
more distant sound and the customer tends to like things in a little closer,
I find that if I mix things so they feel balanced on the Maggies, the
customers like they way they sound on the big horns.

They aren't the most analytic in the top end, but I really do like the
midrange. You can dramatically hear any small EQ changes. And they are
hard to damage. Those two things are big deals for studio monitors.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

"Pooh Bear" wrote
in message
Mike Rieves wrote:

"David Nebenzahl" wrote in
message
.com...
Pooh Bear spake thus:

CWCunningham wrote:

Well I guess I can add this to the long list of
"facts" that you cannot prove,

That amplifier power ratings are measured into
resistive loads ?

As a pro-audio designer ( with a sub-specialisation in
high power amplification )
for ~ 30 yrs I can assure you it's the case.

Please clarify; do you mean *all* amplifiers, or only
(presumably) lower-quality home hi-fi ones? Mikey is
saying that "they" (meaning guys like you, I guess)
design home hi-fi equipment using resistive loads,
while "pro" equipment (studio amps) get the royal
treatment with reactive loads. What say you to this?

That is NOT what I said at all, I said amps are rated,
that is the measurements for their published specs are
made, using resistive loads. NOTHING was said about them
being designed using resistive loads, except by you.
Read what is written before posting!

Reactive loads may be used in testing an amplifier being
designed but not for speccing it.

One simple reason for this is that reactance doesn't
dissipate energy so you couldn't use *watts* for the
spec. This is in fact an interesting point. It would in
fact be somewhat smarter to spec amplifers in terms of
*voltage* output but it would be a task and a half to
change decades of established practice.

Speccing amps in volt-amps would be even more to the point.

"David Nebenzahl" wrote in message
.com

Fine; then design a load that is a combination of
resistances and reactances. Is that so hard to do?

Of course. But couldn't you take an average of real
speakers' values, and use this for the dummy load? It
wouldn't *exactly* match any particular speaker design,
but it should be a hell of a lot better than using a
purely resistive load, no?

There are a number of loads for testing power amps that simulate real-world
speakers. Here's one:

http://www.pcabx.com/product/amplifiers/index.htm

"Pooh Bear" wrote
in message
Mike Rieves wrote:.

3. speaker reactance actually changes according to its
sonic environment, so how is one to know the reactance
of his speakers in his room?.

Does the *reactive component* actually change ? Certainly
the overall impedance does.

Neither changes appreciably unless you do something crazy like face the
speaker into a wall or block a port.

Arny Krueger wrote:

"Pooh Bear" wrote
in message
snip
One simple reason for this is that reactance doesn't
dissipate energy so you couldn't use *watts* for the
spec. This is in fact an interesting point. It would in
fact be somewhat smarter to spec amplifers in terms of
*voltage* output but it would be a task and a half to
change decades of established practice.


Speccing amps in volt-amps would be even more to the point.



???

Aren't volt-amps and watts purt near the same thing?

--
Les Cargill

Les Cargill wrote:
Arny Krueger wrote:

"Pooh Bear" wrote
in message

snip

One simple reason for this is that reactance doesn't
dissipate energy so you couldn't use *watts* for the
spec. This is in fact an interesting point. It would in
fact be somewhat smarter to spec amplifers in terms of
*voltage* output but it would be a task and a half to
change decades of established practice.



Speccing amps in volt-amps would be even more to the point.


???

Aren't volt-amps and watts purt near the same thing?

--
Les Cargill

Purt-near, but not always.
Volt-Amps are just what they sound like. The multiple of
volts and amps ignoring any phase difference. Watts, on the other
hand are the multiple of just the "in-phase" components of voltage
and current, and so is always less than or equal to volt-amps.
For some reason, this is the only thing that stuck with me from my only
EE class. It came back to haunt me when I was selling computer
equipment and had to help people size UPS devices.

Monkey Pi

"Les Cargill" wrote in message

Arny Krueger wrote:

"Pooh Bear"
wrote in message
snip

One simple reason for this is that reactance doesn't
dissipate energy so you couldn't use *watts* for the
spec. This is in fact an interesting point. It would in
fact be somewhat smarter to spec amplifers in terms of
*voltage* output but it would be a task and a half to
change decades of established practice.

Speccing amps in volt-amps would be even more to the
point.

???

Aren't volt-amps and watts purty near the same thing?

No, volt-amps and watts can be vastly different numbers. An ideal reactive
load dissipates no power, so the volt-amps delivered to a reactive load
could be lots while the watts could very small.

"Pooh Bear" wrote in message
...


Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so
how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance
does.


Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance. I suppose that
the term "reactive component" could be taken to mean something besides the
reactance, so maybe that wasn't a good way to put it, but the reactance
definitely does change. :-)

"Scott Dorsey" wrote in message
...
Mike Rieves wrote:
Scott Dorsey writes:
As for me, I'm still using the Magnapans, which are just frighteningly
close to a 6 ohm resistive load.

Magnepans can deliver great sonic realism in the right room with the
right
source material, but they just aren't very good studio monitor speakers.
:-)

Strange, I've been mixing on them for about twenty years now and I've been
very happy. I like the distant presentation a lot; since I tend to like a
more distant sound and the customer tends to like things in a little
closer,
I find that if I mix things so they feel balanced on the Maggies, the
customers like they way they sound on the big horns.

They aren't the most analytic in the top end, but I really do like the
midrange. You can dramatically hear any small EQ changes. And they are
hard to damage. Those two things are big deals for studio monitors.
--scott


But you don't use them for near field monitors, do you? :-) I could see
using Magnepans in a great room for far field or main monitors, but not
without a good pair of nearfields for the fine work. Maggies do have awesome
midrange clarity.

Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance. I suppose that
the term "reactive component" could be taken to mean something besides the
reactance, so maybe that wasn't a good way to put it, but the reactance
definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your
pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like
push the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.


--
I hope that in a few years it [Wikipedia] will be so bloated that it
will simply disintegrate, because I can't stand the thought that this
thing might someday actually be used as a serious reference source.
Because in its current form, it's not to be taken seriously at all.

- Horst Prillinger (see
http://homepage.univie.ac.at/horst.p...06/000623.html)

"David Nebenzahl" wrote in message
.com...
Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic
environment, so how is one to know the reactance of his speakers in his
room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance. I suppose
that the term "reactive component" could be taken to mean something
besides the reactance, so maybe that wasn't a good way to put it, but the
reactance definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your
pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like push
the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.

That's easy, in simple terms, the air in the room acts as a spring (the
smaller the room, the more powerful the spring) which acts on the speaker
cone's motion, which in turn has an effect on the voice coil movement, which
has an effect on the electromotive mechanism. This is reflected back to the
amp output as an impedance change. Standing waves in the room may also have
an affect as well.
This isn't something made up for this news group, that fact that the room
affects a speaker's overall reactance has been known for many years. Anyone
who knows anything about speakers knows that the enclosure the speaker is
installed in has a major effect on a speaker's characteristics, including
impedance, but some may not realize that the room is actually another
enclosure on the other side of the speaker, and it has an effect as well.
Typically, the back of the speaker cone is loading into the enclosure and
the front of the cone is loading into the room.

Les Cargill wrote:

Arny Krueger wrote:

"Pooh Bear" wrote
in message
snip
One simple reason for this is that reactance doesn't
dissipate energy so you couldn't use *watts* for the
spec. This is in fact an interesting point. It would in
fact be somewhat smarter to spec amplifers in terms of
*voltage* output but it would be a task and a half to
change decades of established practice.


Speccing amps in volt-amps would be even more to the point.


???

Aren't volt-amps and watts purt near the same thing?

Watts are real - VA include the imaginary component which could be as large as
100%.

Graham

David Nebenzahl wrote:

Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance.

Actually since a speaker is a motor that does real work, when suitably loaded I
would expect to see changes in the resistive component too.


I suppose that
the term "reactive component" could be taken to mean something besides the
reactance, so maybe that wasn't a good way to put it, but the reactance
definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your
pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like
push the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.

I measured a driver that I initially stupidly placed face down on the bench. The
impedance curve was *very* different to free air. Presumably in a cabinet it would
be somewhere inbetween.

Graham

"Pooh Bear" wrote in message
...
|
|
| CWCunningham wrote:
|
| You've pegged my bull**** meter again. You continue with the claim that I
| encourage you to back up in any meaningful way, but then, as if to add
credence
| to your claim, you state that the FTC has set rules for how Power amps are
to be
| tested. I have to laugh because there couldn't be a less qualified standards
| comittee than the FTC when it comes to EE, nor could there be an industry
less
| deserving of government oversight. But my laughter is likely shortlived
since
| this is a claim you will have little trouble documenting with an easy web
| reference ... I look forward to being made a fool in this regard.
|
| You could try the FTC !
|
| Mike is 100% correct. The FTC introduced this *regulation* in fact to prevent
abuse
| of such ratings as *music power*.
|
100% ... I guess that depends on your rating method, I suppose if your driving a
1khz sine wave through a single channel into one ohm at 10% clipping, the FTC
might let you get away with that (as long as your THD specs are also at 10%
clipping).

Mike states, "Back when the FTC specified how power amps
were to be rated, resistive dummy loads were specified in the rules."

But when I try to find any FTC language that supports that claim, all I find is
the term "impedence", there is no discussion at all of dummy loads. This is
based on the published work of the FTC available on the web which pertains to
the recent amendment and discussions from interested parties. I cannot find the
full text of 16CFR part 432 on the web, and since they want too much money for
copies (as do other standards bodies), I can't say with a certainty that there
isn't verbiage somewhere within the "amplifier rule" to this effect. Here's
where you can be enlightening. I mean you're in the industry with thirty years
experience ... you can go pull CFR volume sixteen right off the shelf and quote
the relevant text both pre and post amendment. If it's in there, it's in there,
and I'm interested, but even if it is, that doesn't make Mike 100% right.

Let's talk about what the "amplifier rule" is not. It's not a law that defines
how amplifiers will be specced, in fact, the FTC couldn't care less how a
manufacturer measures their products. Any manufacturer is free to measure their
products in any way that pleases them and use those measurements in (almost) any
way they see fit. If they create junk with exaggereated specs, independent test
labs will call them on it, so they're not going to fool the type of people who
frequent these groups. Reputable manufacturers will spec their products honestly
and modestly, and gain a reputation for exceeding their published specs.

So what is the amplifier rule good for then? The only time the FTC cares at all
about specs is when you advertise within the USA, *and* you make any reference
to "power", once you do that, you have to comply with the amplifier rule which
gives requirements for how that power will be measured so that anyone who relies
on advertisements to select between products will be able to compare power
ratings on an "apples to apples" basis. Only power ... not freq response, not
signal to noise, not [place any parameter here].

Mike states,
"All of them rate their amps when driving purely resistive loads, and
those are the ratings they publish, they are the ratings listed in the
owners' manuals and in the advertisements for the product".

This is actually false.
Consider http://www.soundstage.com/gettingtec...ical200503.htm

| It's called the 'amplifier rule' btw and was recently updated to take account
of
| recent new product categories such as powered/active speakers.
| http://www.ftc.gov/opa/2000/12/amprule.htm
|
Note in the provided link how the engineers at NAD describe the ammendments as a
loosening of the rules such that an amp will spec more powerful than it could
under the original rules.

--
CWC
============================
It's not that nice guys finish last,
They have a whole different notion
where the finish line is.
============================

"Pooh Bear" wrote in message
...
| *All* amplifiers are *specced* into resistive loads.
|
Since you work in the industry, can you provide a part number for one of these
resistive loads? I just want to check the manufacturer's spec sheet, since I'm
sure you know that any precision resistor that can dissipate more than two watts
is probably a coil inside a heatsink ... it would be interesting to get the
inductance spec for a 100W device.

--
CWC
============================
It's not that nice guys finish last,
They have a whole different notion
where the finish line is.
============================

Pooh Bear spake thus:

David Nebenzahl wrote:

Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance.

Actually since a speaker is a motor that does real work, when suitably loaded I
would expect to see changes in the resistive component too.

I suppose that the term "reactive component" could be taken to mean
something besides the reactance, so maybe that wasn't a good way
to put it, but the reactance definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your
pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like
push the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.

I measured a driver that I initially stupidly placed face down on the bench. The
impedance curve was *very* different to free air. Presumably in a cabinet it would
be somewhere inbetween.

Are you being intentionally obtuse, or just inadvertently so?

Yes, of course: I already said what you just did, as did someone else. A
driver in free air will behave differently from one face-down on a bench
from one in a cabinet. Nobody's disputing that.

What, again, pins my bull**** meter is Mikey's assertion that speaker
impedance/reactance changes in accordance with something he calls the
"sonic environment", which, since it hasn't been defined, I can only
attribute to things like the arrangement of furniture in a listening
room, f'rinstance, or maybe the type of drapes on the windows. None of
which, I assert, has any appreciable, measurable effect on
impedance/reactance. What you're talking about are gross changes in
configuration that naturally would affect this greatly. I'm talking
about his "pixie dust" assertions.


--
I hope that in a few years it [Wikipedia] will be so bloated that it
will simply disintegrate, because I can't stand the thought that this
thing might someday actually be used as a serious reference source.
Because in its current form, it's not to be taken seriously at all.

- Horst Prillinger (see
http://homepage.univie.ac.at/horst.p...06/000623.html)

"Pooh Bear" wrote in message
...
|
|
| CWCunningham wrote:
|
| Well I guess I can add this to the long list of "facts" that you cannot
prove,
|
| That amplifier power ratings are measured into resistive loads ?
|
| As a pro-audio designer ( with a sub-specialisation in high power
amplification )
| for ~ 30 yrs I can assure you it's the case.
|
But you can only speak for yourself, I can assure you it's not absolutely the
case. And while you're here, what's wrong with measuring into a resistive load?
And if there's a qualitative reason that it's "not good" to measure into a
resistive load, why do you do it that way?

--
CWC
============================
It's not that nice guys finish last,
They have a whole different notion
where the finish line is.
============================

CWCunningham wrote:

"Pooh Bear" wrote in message
...
|
|
| CWCunningham wrote:
|
| You've pegged my bull**** meter again. You continue with the claim that I
| encourage you to back up in any meaningful way, but then, as if to add
credence
| to your claim, you state that the FTC has set rules for how Power amps are
to be
| tested. I have to laugh because there couldn't be a less qualified standards
| comittee than the FTC when it comes to EE, nor could there be an industry
less
| deserving of government oversight. But my laughter is likely shortlived
since
| this is a claim you will have little trouble documenting with an easy web
| reference ... I look forward to being made a fool in this regard.
|
| You could try the FTC !
|
| Mike is 100% correct. The FTC introduced this *regulation* in fact to prevent
abuse
| of such ratings as *music power*.

100% ... I guess that depends on your rating method, I suppose if your driving a
1khz sine wave through a single channel into one ohm at 10% clipping, the FTC
might let you get away with that (as long as your THD specs are also at 10%
clipping).

I'd have to check but IIRC the FTC regs don't allow power to be rated at 10% THD.


Mike states, "Back when the FTC specified how power amps
were to be rated, resistive dummy loads were specified in the rules."

But when I try to find any FTC language that supports that claim, all I find is
the term "impedence", there is no discussion at all of dummy loads. This is
based on the published work of the FTC available on the web which pertains to
the recent amendment and discussions from interested parties. I cannot find the
full text of 16CFR part 432 on the web, and since they want too much money for
copies (as do other standards bodies), I can't say with a certainty that there
isn't verbiage somewhere within the "amplifier rule" to this effect. Here's
where you can be enlightening. I mean you're in the industry with thirty years
experience ... you can go pull CFR volume sixteen right off the shelf and quote
the relevant text both pre and post amendment. If it's in there, it's in there,
and I'm interested, but even if it is, that doesn't make Mike 100% right.

There's a simple explanation. If the load is reactive the dissipation in the
amplifer's output stage is *higher*, so no sensible manufacturere would use anything
other than a resistive load in order to get the best figures.


Let's talk about what the "amplifier rule" is not. It's not a law that defines
how amplifiers will be specced, in fact, the FTC couldn't care less how a
manufacturer measures their products.

Well, yes it is at least in part because of the thermal considerations. That's
exactly why it was introduced. It also outlaws music power and peak music power
ratings.


Any manufacturer is free to measure their
products in any way that pleases them and use those measurements in (almost) any
way they see fit. If they create junk with exaggereated specs, independent test
labs will call them on it, so they're not going to fool the type of people who
frequent these groups.

They can fool the general public though. Which is the reason for the rule.


Reputable manufacturers will spec their products honestly
and modestly, and gain a reputation for exceeding their published specs.

It was precisely to stop *disreputable* manufacturers.

So what is the amplifier rule good for then?

One standardised method that all have to play by and it's worked too.

The only time the FTC cares at all
about specs is when you advertise within the USA, *and* you make any reference
to "power", once you do that, you have to comply with the amplifier rule which
gives requirements for how that power will be measured so that anyone who relies
on advertisements to select between products will be able to compare power
ratings on an "apples to apples" basis. Only power ... not freq response, not
signal to noise, not [place any parameter here].

Do you really think it makes sense to publish different specs for the USA only ?

Graham

CWCunningham wrote:

"Pooh Bear" wrote in message
...
| *All* amplifiers are *specced* into resistive loads.
|
Since you work in the industry, can you provide a part number for one of these
resistive loads? I just want to check the manufacturer's spec sheet, since I'm
sure you know that any precision resistor that can dissipate more than two watts
is probably a coil inside a heatsink ... it would be interesting to get the
inductance spec for a 100W device.

I actually measured one once. It had about 4uH of inductance. Purists would use
non-inductively wound Rs but I'm happy to accept that 4 uH !

http://uk.farnell.com/jsp/endecaSear...tk=g ensearch

Graham

David Nebenzahl wrote:

Pooh Bear spake thus:

David Nebenzahl wrote:

Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance.

Actually since a speaker is a motor that does real work, when suitably loaded I
would expect to see changes in the resistive component too.

I suppose that the term "reactive component" could be taken to mean
something besides the reactance, so maybe that wasn't a good way
to put it, but the reactance definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your
pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like
push the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.

I measured a driver that I initially stupidly placed face down on the bench. The
impedance curve was *very* different to free air. Presumably in a cabinet it would
be somewhere inbetween.

Are you being intentionally obtuse, or just inadvertently so?

Yes, of course: I already said what you just did, as did someone else. A
driver in free air will behave differently from one face-down on a bench
from one in a cabinet. Nobody's disputing that.

What, again, pins my bull**** meter is Mikey's assertion that speaker
impedance/reactance changes in accordance with something he calls the
"sonic environment", which, since it hasn't been defined, I can only
attribute to things like the arrangement of furniture in a listening
room, f'rinstance, or maybe the type of drapes on the windows. None of
which, I assert, has any appreciable, measurable effect on
impedance/reactance.

I didn't think he meant that. I took it to mean size of room etc.

What you're talking about are gross changes in
configuration that naturally would affect this greatly. I'm talking
about his "pixie dust" assertions.

Since you accept that large changes exist due to loading you have to accept also that
small changes must also exist.

Graham

CWCunningham wrote:

"Scott Dorsey" wrote in message
...
| David Nebenzahl wrote:

| Fine; then design a load that is a combination of resistances and
| reactances. Is that so hard to do?
|
| It's easy to do. But which speaker does it represent? There are a couple
| ISO standards that are out there for testing purposes, but neither one of
| them represent any particular speaker.
|
| Look at the impedance curves of an Apogee Scintilla, a Bose 901, and a
| Quad ESL57. They all look totally different from one another. They all
| present totally different loads to the amp and move the poles and zeros
| around to different places. Which one do you use?

This is a straw man argument.

No it isn't.

I'll grant that different speakers have different
characteristics and so one model can't be all things. But it doesn't follow that
no model is better than one model ... and why not use two or three?

Or ten or twenty or a hundred ?

I'm sure the
answer is economics, but at some point, a decision has to be made, and I'm sure
that in companies that care about quality, that decision was made long ago.

It's less about economics than simple practicality. If you used the worst example of
reactive behaviour as a test standard then many otherwise perfectly good amplifiers
( when used 'typically' ) would produce dismal results.

It's down to the user to be aware of this.

You could say something similar about fuel octane. Would you expect a Ferrari to run
well on 85 octane ?

Either resistive loads produce quality results, or we test into reactive loads
that are reasonably representitive of our target impedence. Granted such a
reactive load may only mimic a small range of speakers ... but the results are
meaningful even when driving a speaker that presents a different load ... all
over the map is a pretty finite space, and a quality manufacturer takes the time
to do their job well.

I wish you well convincing the AES etc.

Graham

CWCunningham wrote:

"Pooh Bear" wrote in message
...
|
|
| CWCunningham wrote:
|
| Well I guess I can add this to the long list of "facts" that you cannot
prove,
|
| That amplifier power ratings are measured into resistive loads ?
|
| As a pro-audio designer ( with a sub-specialisation in high power
amplification )
| for ~ 30 yrs I can assure you it's the case.
|
But you can only speak for yourself,

I can speak for my colleauges too. It's known and accepted in the industry
*universally*.

I can assure you it's not absolutely the case.

You're talking nonsense.

And while you're here, what's wrong with measuring into a resistive load?
And if there's a qualitative reason that it's "not good" to measure into a
resistive load, why do you do it that way?

Because it's the only possible standard. Once you get into xR +/- yj you can have
multiple combinations of x and y and + or -. Not to mention the test frequency too.

Graham

Pooh Bear spake thus:

David Nebenzahl wrote:

Pooh Bear spake thus:

David Nebenzahl wrote:

Mike Rieves spake thus:

"Pooh Bear" wrote in message
...

Mike Rieves wrote:.

3. speaker reactance actually changes according to its sonic environment,
so how is one to know the reactance of his speakers in his room?.

Does the *reactive component* actually change ? Certainly the overall
impedance does.

Since the resistive component doesn't change, if the inpedance changes
(and it does), the change would have to be in the reactance.

Actually since a speaker is a motor that does real work, when suitably loaded I
would expect to see changes in the resistive component too.

I suppose that the term "reactive component" could be taken to mean
something besides the reactance, so maybe that wasn't a good way
to put it, but the reactance definitely does change. :-)

And how, pray tell, does the impedance (or reactance, or both; take your

pick) change according to its "sonic environment"?

As someone else stated here, it's *maybe* conceiveable that a speaker's
reactance/impedance *might* change if one did something drastic, like
push the speaker up against a wall.

Any assertions that the impedance/reactance changes appreciably due to
change in the "sonic environment" (whatever the ****** that is) are just
audiophool pixie-dust fantasies. Like maybe small changes in barometric
pressure? Relative humidity?

Come on; this ought to be good. Explain away. Get specific.

I measured a driver that I initially stupidly placed face down on the bench. The
impedance curve was *very* different to free air. Presumably in a cabinet it would
be somewhere inbetween.

Are you being intentionally obtuse, or just inadvertently so?

Yes, of course: I already said what you just did, as did someone else. A
driver in free air will behave differently from one face-down on a bench
from one in a cabinet. Nobody's disputing that.

What, again, pins my bull**** meter is Mikey's assertion that speaker
impedance/reactance changes in accordance with something he calls the
"sonic environment", which, since it hasn't been defined, I can only
attribute to things like the arrangement of furniture in a listening
room, f'rinstance, or maybe the type of drapes on the windows. None of
which, I assert, has any appreciable, measurable effect on
impedance/reactance.

I didn't think he meant that. I took it to mean size of room etc.

All right, I'll accept that for the sake of argument. Still wouldn't
make any difference so far as the *measurable* impedance/reactance of
the speaker goes; the travel of the cone, air resistance, damping, etc.,
isn't gonna change just because the room has a different shape.

What you're talking about are gross changes in
configuration that naturally would affect this greatly. I'm talking
about his "pixie dust" assertions.

Since you accept that large changes exist due to loading you have to accept also that
small changes must also exist.

Yes; small as in inconsequential. As in immeasurable. As in irrelevant.


--
I hope that in a few years it [Wikipedia] will be so bloated that it
will simply disintegrate, because I can't stand the thought that this
thing might someday actually be used as a serious reference source.
Because in its current form, it's not to be taken seriously at all.

- Horst Prillinger (see
http://homepage.univie.ac.at/horst.p...06/000623.html)

Pooh Bear spake thus:

CWCunningham wrote:

"Pooh Bear" wrote in message
...
|
| CWCunningham wrote:
|
| Well I guess I can add this to the long list of "facts" that you cannot
prove, That amplifier power ratings are measured into resistive loads ?
|
| As a pro-audio designer ( with a sub-specialisation in high power
amplification ) for ~ 30 yrs I can assure you it's the case.
|
But you can only speak for yourself,

I can speak for my colleauges too. It's known and accepted in the industry
*universally*.

I can assure you it's not absolutely the case.

You're talking nonsense.

And while you're here, what's wrong with measuring into a resistive load?
And if there's a qualitative reason that it's "not good" to measure into a
resistive load, why do you do it that way?

Because it's the only possible standard. Once you get into xR +/- yj you can have
multiple combinations of x and y and + or -. Not to mention the test frequency too.

So I'm curious what you think of Arny Krueger's reactive dummy load:
http://www.pcabx.com/product/amplifiers/index.htm. Nobody has commented
on this yet.

Here's a real-world example of just what I was proposing: a load that
combined resistance and reactance to mimic an (average) real-world load
(i.e., a speaker). With an admittedly arbitrarily-chosen topology, if
that's the right word; hey, you have to choose one.


--
I hope that in a few years it [Wikipedia] will be so bloated that it
will simply disintegrate, because I can't stand the thought that this
thing might someday actually be used as a serious reference source.
Because in its current form, it's not to be taken seriously at all.

- Horst Prillinger (see
http://homepage.univie.ac.at/horst.p...06/000623.html)

"Pooh Bear" wrote in message
...
|
|
| CWCunningham wrote:
|
| "Scott Dorsey" wrote in message
| ...
| | David Nebenzahl wrote:
|
| | Fine; then design a load that is a combination of resistances and
| | reactances. Is that so hard to do?
| |
| | It's easy to do. But which speaker does it represent? There are a couple
| | ISO standards that are out there for testing purposes, but neither one of
| | them represent any particular speaker.
| |
| | Look at the impedance curves of an Apogee Scintilla, a Bose 901, and a
| | Quad ESL57. They all look totally different from one another. They all
| | present totally different loads to the amp and move the poles and zeros
| | around to different places. Which one do you use?
|
| This is a straw man argument.
|
| No it isn't.
|
| I'll grant that different speakers have different
| characteristics and so one model can't be all things. But it doesn't follow
that
| no model is better than one model ... and why not use two or three?
|
| Or ten or twenty or a hundred ?
|
There's that straw man ... oh but you said ... oh

| I'm sure the
| answer is economics, but at some point, a decision has to be made, and I'm
sure
| that in companies that care about quality, that decision was made long ago.
|
| It's less about economics than simple practicality. If you used the worst
example of
| reactive behaviour as a test standard then many otherwise perfectly good
amplifiers
| ( when used 'typically' ) would produce dismal results.
|
So why use the worst example ... are you trying to weave straw into something of
value?

| It's down to the user to be aware of this.
|
| You could say something similar about fuel octane. Would you expect a Ferrari
to run
| well on 85 octane ?
|
No, but on straw of this quality, I'm sure it will fly, in fact an enzo ferrari
is said to be able to travel upside down with it's dynamic downforce, but I'm
sure they test using straw.

| Either resistive loads produce quality results, or we test into reactive
loads
| that are reasonably representitive of our target impedence. Granted such a
| reactive load may only mimic a small range of speakers ... but the results
are
| meaningful even when driving a speaker that presents a different load ...
all
| over the map is a pretty finite space, and a quality manufacturer takes the
time
| to do their job well.
|
| I wish you well convincing the AES etc.
|
Quality doesn't care about the simple stuff.

--
CWC
============================
It's not that nice guys finish last,
They have a whole different notion
where the finish line is.
============================

David Nebenzahl wrote:

Pooh Bear spake thus:

CWCunningham wrote:

And while you're here, what's wrong with measuring into a resistive load?
And if there's a qualitative reason that it's "not good" to measure into a
resistive load, why do you do it that way?

Because it's the only possible standard. Once you get into xR +/- yj you can have
multiple combinations of x and y and + or -. Not to mention the test frequency too.

So I'm curious what you think of Arny Krueger's reactive dummy load:
http://www.pcabx.com/product/amplifiers/index.htm. Nobody has commented
on this yet.

It's just *an example* of what's possible.

Here's a real-world example of just what I was proposing: a load that
combined resistance and reactance to mimic an (average) real-world load
(i.e., a speaker). With an admittedly arbitrarily-chosen topology, if
that's the right word; hey, you have to choose one.

All reactive loads will be arbitrary. Resistive ones aren't. Hence that's why they're
used. Common playing field.

Graham

CWCunningham wrote:

"Pooh Bear" wrote in message
...

| It's less about economics than simple practicality. If you used the worst
example of
| reactive behaviour as a test standard then many otherwise perfectly good
amplifiers
| ( when used 'typically' ) would produce dismal results.

So why use the worst example ... are you trying to weave straw into something of
value?

I don't understand what you mean by that.

Withoutnusing a 'worst example' what would be the point at all ?

| It's down to the user to be aware of this.
|
| You could say something similar about fuel octane. Would you expect a Ferrari
to run
| well on 85 octane ?

No, but on straw of this quality, I'm sure it will fly, in fact an enzo ferrari
is said to be able to travel upside down with it's dynamic downforce, but I'm
sure they test using straw.

You're being silly.

Graham

On Thu, 01 Jun 2006 05:21:06 +0100, Pooh Bear
wrote:

Aren't volt-amps and watts purt near the same thing?

Watts are real - VA include the imaginary component which could be as large as
100%.

Serious question:

If a real voltage is pushing real amps but wattage is less than V X A,
where is the power going?

Laurence Payne wrote:

On Thu, 01 Jun 2006 05:21:06 +0100, Pooh Bear
wrote:

Aren't volt-amps and watts purt near the same thing?

Watts are real - VA include the imaginary component which could be as large as
100%.

Serious question:

If a real voltage is pushing real amps but wattage is less than V X A,
where is the power going?

It's being dissipated in the output devices. Loads that are 100% reactive dissipate
no power. So, for example, driving an amplifier into a perfect capacitor with no
losses will deliver both volts and amps to the load but no power. The output stage
of the ampliifer will get hot though.

In power distribution systems this difference between W and VA is called 'power
factor'. W = real power VA = 'apparent power'.

Graham

Pooh Bear wrote:

David Nebenzahl wrote:

Pooh Bear spake thus:

David Nebenzahl wrote:

What, again, pins my bull**** meter is Mikey's assertion that speaker
impedance/reactance changes in accordance with something he calls the
"sonic environment", which, since it hasn't been defined, I can only
attribute to things like the arrangement of furniture in a listening
room, f'rinstance, or maybe the type of drapes on the windows. None of
which, I assert, has any appreciable, measurable effect on
impedance/reactance.
I didn't think he meant that. I took it to mean size of room etc.
All right, I'll accept that for the sake of argument. Still wouldn't
make any difference so far as the *measurable* impedance/reactance of
the speaker goes; the travel of the cone, air resistance, damping, etc.,
isn't gonna change just because the room has a different shape.

Not so much shape as such.

What you're talking about are gross changes in
configuration that naturally would affect this greatly. I'm talking
about his "pixie dust" assertions.
Since you accept that large changes exist due to loading you have to accept also that
small changes must also exist.
Yes; small as in inconsequential. As in immeasurable. As in irrelevant.

Possibly not entirely irrelevant actually.

There are also other ways in which a room and speaker interact too. A room and speaker
positioning can have a large influence.

It has big influence on the sound from somewhere in the room but none
at all on the dynamics of the motion of the cone and whatever is in
the port (air or passive radiator) in response to a signal. The
coupling of a speaker to its cabinet loading totally swamps its
coupling to the room.


Bob
--

"Things should be described as simply as possible, but no simpler."

A. Einstein

"Mike Rieves" wrote in message


That's easy, in simple terms, the air in the room acts
as a spring (the smaller the room, the more powerful the
spring) which acts on the speaker cone's motion, which in
turn has an effect on the voice coil movement, which has
an effect on the electromotive mechanism. This is
reflected back to the amp output as an impedance change.

You're presuming that the speaker works like a mic, which of course it
does - to a degree.

Let's try to roughly quantify this effect.

A typical direct radiator speaker is about 1% efficient as a speaker, which
means that being a symetric device, it is about 1% efficient as a mic. IOW
any effects due to the sound in the room are divided by 100. With some luck
you might even be able to measure the changes on a speaker's impedance curve
due to reflections in the room. BTW, the room isn't 100% efficient either,
so the sound that bounces back at the speaker is itself highly attenuated.

"Pooh Bear" wrote
in message

I measured a driver that I initially stupidly placed face
down on the bench. The impedance curve was *very*
different to free air. Presumably in a cabinet it would
be somewhere inbetween.

So extreme as to be irrelevant to the discussion at hand, unless you call
slapping the driver side of a speaker system up against a wall "typical
speaker placement in a room".

"Laurence Payne" lpayne1NOSPAM@dslDOTpipexDOTcom wrote in
message
On Thu, 01 Jun 2006 05:21:06 +0100, Pooh Bear
wrote:

Aren't volt-amps and watts purt near the same thing?

Watts are real - VA include the imaginary component
which could be as large as 100%.

Serious question:

If a real voltage is pushing real amps but wattage is
less than V X A, where is the power going?

The power was never drawn from the source.

In the real world, one finds that speaker loads cause power amps to draw a
ton less current from the wall than a resistive load with the same nominal
impedance.

Pooh Bear spake thus:

David Nebenzahl wrote:

Pooh Bear spake thus:

David Nebenzahl wrote:

What, again, pins my bull**** meter is Mikey's assertion that speaker
impedance/reactance changes in accordance with something he calls the
"sonic environment", which, since it hasn't been defined, I can only
attribute to things like the arrangement of furniture in a listening
room, f'rinstance, or maybe the type of drapes on the windows. None of
which, I assert, has any appreciable, measurable effect on
impedance/reactance.

I didn't think he meant that. I took it to mean size of room etc.

All right, I'll accept that for the sake of argument. Still wouldn't
make any difference so far as the *measurable* impedance/reactance of
the speaker goes; the travel of the cone, air resistance, damping, etc.,
isn't gonna change just because the room has a different shape.

Not so much shape as such.

What you're talking about are gross changes in
configuration that naturally would affect this greatly. I'm talking
about his "pixie dust" assertions.

Since you accept that large changes exist due to loading you have to accept also that
small changes must also exist.

Yes; small as in inconsequential. As in immeasurable. As in irrelevant.

Possibly not entirely irrelevant actually.

There are also other ways in which a room and speaker interact too. A room and speaker
positioning can have a large influence.

ON SOUND, yes; on impedance/reactance? I don't think so.


--
Any system of knowledge that is capable of listing films in order of
use of the word "****" is incapable of writing a good summary and
analysis of the Philippine-American War. And vice-versa. This is an
inviolable rule.

- Matthew White, referring to Wikipedia on his WikiWatch site
(http://users.erols.com/mwhite28/wikiwoo.htm)

"CWCunningham" charlesw-at-blackfoot.net wrote in message


I have to laugh because there
couldn't be a less qualified standards comittee than the
FTC when it comes to EE,

The FTC rule was actually formulated by the IHFM
(Institute of High Fidelity Manufacturers) which was the leading industry
trade group at the time. The IHFM was later absorbed by the EIA.

The IHFM group that formulated the rule was populated by technical folks who
were technical leaders in audio at that time, people like Julian Hirsch and
Norman Crowhurst.

nor could there be an industry less deserving of government oversight.

Things were pretty wild and wooly at the time. For example, it was common to
advertise amplifier power "+/- 1 dB" which allowed taking on about 20% that
was pure speculation.

"David Nebenzahl" wrote in message
.com

There are also other ways in which a room and speaker
interact too. A room and speaker positioning can have a
large influence.

ON SOUND, yes; on impedance/reactance? I don't think so.

Real world measurements bear this out. Once you put a speaker in a box and
separate that box from nearby walls by the largest dimension of that box or
even a fraction of it, its impedance curve is essentially unchanged.

Mike Rieves wrote:

But you don't use them for near field monitors, do you? :-) I could see
using Magnepans in a great room for far field or main monitors, but not
without a good pair of nearfields for the fine work. Maggies do have awesome
midrange clarity.

No, I never really got the whole near-field thing. When near field monitoring
started coming in, it came in with a huge amount of hype about eliminating
room problems and how you wouldn't have to treat the room any more. Needless
to say this turned out to be false. But in the eighties I saw folks just
going berserk over the idea and I never felt like I ever liked working
thhat way.

I'll occasionally track on nearfields and I use them in the truck all
the time because there just isn't room for a decent monitoring system.
But for the most part I don't really like mixing on nearfields at all
if there is a good alternative.

Then again, I learned to mix on Altec 604s. Thank God that those days
are over.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

CWCunningham charlesw-at-blackfoot.net wrote:
This is a straw man argument.I'll grant that different speakers have different
characteristics and so one model can't be all things. But it doesn't follow that
no model is better than one model ... and why not use two or three? I'm sure the
answer is economics, but at some point, a decision has to be made, and I'm sure
that in companies that care about quality, that decision was made long ago.

Lots of folks who do reviews DO use several different models.

The problem is that the people who write the spec sheet are the manufacturers.
They want their products to look good. If you give them a choice of several
different loads, they will pick the one that makes their product look good,
and THEN you won't be able to compare products any more.

Take noise floor measurements on microphones as an example. Everybody does
them differently, they all cite a different standard, and they all pick the
standard that makes their product look best (except for the people who make
the numbers up out of whole cloth).

Either resistive loads produce quality results, or we test into reactive loads
that are reasonably representitive of our target impedence. Granted such a
reactive load may only mimic a small range of speakers ... but the results are
meaningful even when driving a speaker that presents a different load ... all
over the map is a pretty finite space, and a quality manufacturer takes the time
to do their job well.

Resistive loads tell you something valuable and useful about the amp.
It doesn't tell you everything you need to know, but no single scalar
number is going to.

I will say that before the FTC standards appeared, one manufacturer did
do all the testing for their data sheet numbers into a slightly inductive
load... because the amplifier would rapidly become unstable into a load
that wasn't slightly inductive. Needless to say reviewers trying to
reproduce their numbers into dummy loads had some trouble doing so. The
vendor's argument is that most speakers were inductive enough that the amps
wouldn't blow up, and therefore testing them with such a load was okay.
Leonard Feldman's argument was that since you can't guarantee what the end
user will load the amp with, you have to use a generic dummy load.

When Len Feldman is arguing that your numbers are cooked, they are REALLY
cooked.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

"CWCunningham" charlesw-at-blackfoot.net wrote in message
...
"Scott Dorsey" wrote in message
...
| David Nebenzahl wrote:
| Scott Dorsey spake thus:
|
| David Nebenzahl wrote:
|
| You mean to tell us that there's no way to design a dummy load that
| 1) has the same effective reactance as a real load (like a speaker)
and
| 2) can dissipate the amp's power?
|
| No, he's saying that the reactive part doesn't dissipate the power.
Real
| speakers are a combination of resistances and reactances.
|
| Fine; then design a load that is a combination of resistances and
| reactances. Is that so hard to do?
|
| It's easy to do. But which speaker does it represent? There are a
couple
| ISO standards that are out there for testing purposes, but neither one
of
| them represent any particular speaker.
|
| Look at the impedance curves of an Apogee Scintilla, a Bose 901, and a
| Quad ESL57. They all look totally different from one another. They all
| present totally different loads to the amp and move the poles and zeros
| around to different places. Which one do you use?
|
This is a straw man argument.I'll grant that different speakers have
different
characteristics and so one model can't be all things. But it doesn't
follow that
no model is better than one model ... and why not use two or three? I'm
sure the
answer is economics, but at some point, a decision has to be made, and I'm
sure
that in companies that care about quality, that decision was made long
ago.

Either resistive loads produce quality results, or we test into reactive
loads
that are reasonably representitive of our target impedence. Granted such a
reactive load may only mimic a small range of speakers ... but the results
are
meaningful even when driving a speaker that presents a different load ...
all
over the map is a pretty finite space, and a quality manufacturer takes
the time
to do their job well.

How do you decide what the "target impedance" is? The results would be
meaningless on any speaker other than the "target" system, since every
divergance of reactance from the target value would give a different
response at some frequency.
And how is the end user, who doesn't have a clue about resistance or
reactance, going to decide which set of specs, from which dummy load,
represents his speakers and with what degree of accuracy? Since speaker
reactance is all over the place, no dummy load is likely to accurately
represent any given speaker other than the one it's modeled on. In this case
publishiung specs gaind from using reactive loads would be no more accurate
and much more confusing than the current resistive load rating system.
The only way to gain even fairly accurate specs would be to model a
reactive load for every speaker on the market and publish them all in a
database that the end user could study, and even this wouldn't be anywhere
near perfect because of the room's effect on the speaker.

A reactive load that doesn't represent the one you're using tells you
something
more useful than a resistive load that doesn't represent any speaker at
all.


No, CW, it doesn't, not unless you're an engineer with access to all data
concerning the load and the speaker in question, including enclosure and
room data. The thing about using a resistive load is that it is a general
approximation, as innaccurate in one direction of reactance as it is in the
other, sort of a "middle of the road" approach. Remember that reactance can
be capacitive series or shunt equivalent or inductive series or shunt
equivalent, each going in opposite directions. Current published specs give
little useful specific information but they do give some general
information. You can be pretty sure that an amp rated at 100Wrms will give
more acoustic output from your speakers than will one rated at 10Wrms,
assuming that the rating method is the same, but even that can be iffy, if
the rating methods aren't the same.
Also remember that there are different rating methods currently in use,
making it even more confusing. I've seen amps rated as: "RMS continuous
power ", "RMS music power", "RMS peak power", "peak power", and others, none
of which have much in common. With so many rating methods now in use, adding
more will only increase the confusion, not clarify anything.