Null speaker test

[[email protected]]

Years ago I read an article by Larry Klein IIRC, that describe a way to find
differences in amplifiers. It was called the null speaker test, again IIRC.

Essentially it was a way to hook up 2 amplifiers and a single loudspeaker so
that the speaker would generate sound if there were any differences between
the 2.

Has anyone heard of this test, and if so, can you point me towards a place
that describes the setup in detail?

[AZ Nomad]

On Tue, 11 Oct 2005 22:53:15 GMT, > wrote:


>Years ago I read an article by Larry Klein IIRC, that describe a way to find
>differences in amplifiers. It was called the null speaker test, again IIRC.

>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker so
>that the speaker would generate sound if there were any differences between
>the 2.

>Has anyone heard of this test, and if so, can you point me towards a place
>that describes the setup in detail?


Sounds like hooking a speaker between the (+) post of one amp and the (+)
post of the other amp. Several problems come to mind:
1) won't with with amps that use complementary (push/pull) outputs
2) won't tell you which amp is in error
3) won't run the amps with the load for which they're designed
4) and what's the point of this silly exercise? there are far better ways
to analyze amplifier distortion. Even connecting a scope on the amp's
inputs and outputs will give a more meaningfull display of amplifier
distortion.

[Kalman Rubinson]

On Tue, 11 Oct 2005 22:53:15 GMT, "
> wrote:

>Years ago I read an article by Larry Klein IIRC, that describe a way to find
>differences in amplifiers. It was called the null speaker test, again IIRC.
>
>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker so
>that the speaker would generate sound if there were any differences between
>the 2.
>
>Has anyone heard of this test, and if so, can you point me towards a place
>that describes the setup in detail?

Cannot recall that but I do recall something similar that was advanced
by Hafler. Probably involves inverted polarity feeds to two amps with
the amp running off the two (+) terminals or a summing of the outputs.

Kal

[[email protected]]

"Kalman Rubinson" > wrote in message
...
> On Tue, 11 Oct 2005 22:53:15 GMT, "
> > wrote:
>
>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>find
>>differences in amplifiers. It was called the null speaker test, again
>>IIRC.
>>
>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>so
>>that the speaker would generate sound if there were any differences
>>between
>>the 2.
>>
>>Has anyone heard of this test, and if so, can you point me towards a place
>>that describes the setup in detail?
>
> Cannot recall that but I do recall something similar that was advanced
> by Hafler. Probably involves inverted polarity feeds to two amps with
> the amp running off the two (+) terminals or a summing of the outputs.
>
> Kal
That was the poor man's way of extracting ambient sound before true surround
sound came into being.

The set up I'm trying to get details on, only generated sound if there were
a differnce.
AZ Nomad is correct in that it won't tell you which amp is genrating the
difference, but kn owing if there is any difference would be useful in its
own right.

[Robert Morein]

" > wrote in message
hlink.net...
> Years ago I read an article by Larry Klein IIRC, that describe a way to
find
> differences in amplifiers. It was called the null speaker test, again
IIRC.
>
> Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
so
> that the speaker would generate sound if there were any differences
between
> the 2.
>
> Has anyone heard of this test, and if so, can you point me towards a place
> that describes the setup in detail?
>
Yes, I own a Hafler tester.
With the Hafler bridge, only one amplifier is required.
The Hafler bridge does not accommodate two amplifiers.

[Robert Morein]

"Kalman Rubinson" > wrote in message
...
> On Tue, 11 Oct 2005 22:53:15 GMT, "
> > wrote:
>
> >Years ago I read an article by Larry Klein IIRC, that describe a way to
find
> >differences in amplifiers. It was called the null speaker test, again
IIRC.
> >
> >Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
so
> >that the speaker would generate sound if there were any differences
between
> >the 2.
> >
> >Has anyone heard of this test, and if so, can you point me towards a
place
> >that describes the setup in detail?
>
> Cannot recall that but I do recall something similar that was advanced
> by Hafler. Probably involves inverted polarity feeds to two amps with
> the amp running off the two (+) terminals or a summing of the outputs.
>
> Kal

It consists of a bridge. One input to the bridge is the line-level input to
the amplifier.
The second input to the bridge is the output of the power amplifier, passed
through a voltage divider.
The bridge output consists of the difference signal between the input to the
amplifier, and the output.

Hafler claimed that the XL series of amplifiers exhibited a level of
distortion, when measured in this fashion, 20dB below other extant
amplifiers. The box was to be loaned by dealers so that XL owners could make
a nulling adjustment in the field. The nulling adjustment is a capacitor
that adjusts an ultrasonic peak in the frequency response of the amplifier,
thus providing an open loop compensation for load reactance at high
frequency. The XL design also claims an extremely low phase shift: 1/4
degree from 20Hz to 20kHz.

Reviewers seemed to ignore this feature, preferring to test the amplifier
only at the factory default setting. But if Hafler's claims are to be
believed, his method picked up, as an undifferentiated power, types of
distortion that were or are generally ignored by HM and IM measurements.
This is one reason I am skeptical of claims that amplifier distortion has
been adequately quantified.

[AZ Nomad]

On Tue, 11 Oct 2005 22:53:15 GMT, > wrote:


>Years ago I read an article by Larry Klein IIRC, that describe a way to find
>differences in amplifiers. It was called the null speaker test, again IIRC.

>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker so
>that the speaker would generate sound if there were any differences between
>the 2.

>Has anyone heard of this test, and if so, can you point me towards a place
>that describes the setup in detail?


P.S. learn how to crosspost. Starting multiple independant threads
in different newsgroups is assinine.

[ScottW]

"Kalman Rubinson" > wrote in message
...
> On Tue, 11 Oct 2005 22:53:15 GMT, "
> > wrote:
>
>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>find
>>differences in amplifiers. It was called the null speaker test, again
>>IIRC.
>>
>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>so
>>that the speaker would generate sound if there were any differences
>>between
>>the 2.
>>
>>Has anyone heard of this test, and if so, can you point me towards a place
>>that describes the setup in detail?
>
> Cannot recall that but I do recall something similar that was advanced
> by Hafler. Probably involves inverted polarity feeds to two amps with
> the amp running off the two (+) terminals or a summing of the outputs.
>
> Kal

I don't think you want to invert the feeds. Take two amps set at identical
gain with a common feed, tie the grounds of each amp together (neg speaker
terminal) and connect the positive output of one amp to the + speaker
terminal, the positive of the other amp to the negative speaker terminal.
The speaker will only "play" the difference of the amps.

Seems to me if you invert the polarity to one amp you will always have a
difference.

I can't see how this method would be very revealing. Maybe if you took the
difference signal and ran it through a 3rd amp to get some gain on it.

ScottW

[Kalman Rubinson]

On Tue, 11 Oct 2005 19:25:02 -0700, "ScottW" >
wrote:

>I don't think you want to invert the feeds. Take two amps set at identical
>gain with a common feed, tie the grounds of each amp together (neg speaker
>terminal) and connect the positive output of one amp to the + speaker
>terminal, the positive of the other amp to the negative speaker terminal.
>The speaker will only "play" the difference of the amps.
>
>Seems to me if you invert the polarity to one amp you will always have a
>difference.
>
>I can't see how this method would be very revealing. Maybe if you took the
>difference signal and ran it through a 3rd amp to get some gain on it.

Yes, of course, you are right and I was not thinking it through.

Kal

[Clyde Slick]

"AZ Nomad" > wrote in message
...
> On Tue, 11 Oct 2005 22:53:15 GMT,
> > wrote:
>
>
>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>find
>>differences in amplifiers. It was called the null speaker test, again
>>IIRC.
>
>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>so
>>that the speaker would generate sound if there were any differences
>>between
>>the 2.
>
>>Has anyone heard of this test, and if so, can you point me towards a place
>>that describes the setup in detail?
>
>
> P.S. learn how to crosspost. Starting multiple independant threads
> in different newsgroups is assinine.

Damned if you do, damned if you don't

[[email protected]]

"Robert Morein" > wrote in message
...
>
> "Kalman Rubinson" > wrote in message
> ...
>> On Tue, 11 Oct 2005 22:53:15 GMT, "
>> > wrote:
>>
>> >Years ago I read an article by Larry Klein IIRC, that describe a way to
> find
>> >differences in amplifiers. It was called the null speaker test, again
> IIRC.
>> >
>> >Essentially it was a way to hook up 2 amplifiers and a single
>> >loudspeaker
> so
>> >that the speaker would generate sound if there were any differences
> between
>> >the 2.
>> >
>> >Has anyone heard of this test, and if so, can you point me towards a
> place
>> >that describes the setup in detail?
>>
>> Cannot recall that but I do recall something similar that was advanced
>> by Hafler. Probably involves inverted polarity feeds to two amps with
>> the amp running off the two (+) terminals or a summing of the outputs.
>>
>> Kal
>
> It consists of a bridge. One input to the bridge is the line-level input
> to
> the amplifier.
> The second input to the bridge is the output of the power amplifier,
> passed
> through a voltage divider.
> The bridge output consists of the difference signal between the input to
> the
> amplifier, and the output.
>
> Hafler claimed that the XL series of amplifiers exhibited a level of
> distortion, when measured in this fashion, 20dB below other extant
> amplifiers. The box was to be loaned by dealers so that XL owners could
> make
> a nulling adjustment in the field. The nulling adjustment is a capacitor
> that adjusts an ultrasonic peak in the frequency response of the
> amplifier,
> thus providing an open loop compensation for load reactance at high
> frequency. The XL design also claims an extremely low phase shift: 1/4
> degree from 20Hz to 20kHz.
>
> Reviewers seemed to ignore this feature, preferring to test the amplifier
> only at the factory default setting. But if Hafler's claims are to be
> believed, his method picked up, as an undifferentiated power, types of
> distortion that were or are generally ignored by HM and IM measurements.
> This is one reason I am skeptical of claims that amplifier distortion has
> been adequately quantified.
>
I don't know if this is the process I was trying to remember or not, but the
main reason I mention it is that it would be a very good way to find out if
there's any differnce between two amps. No sound, no difference, no
difference, no upgrade. And you don't need an ABX box.

[[email protected]]

"ScottW" > wrote in message
news:5u_2f.3480$jw6.2919@lakeread02...
>
> "Kalman Rubinson" > wrote in message
> ...
>> On Tue, 11 Oct 2005 22:53:15 GMT, "
>> > wrote:
>>
>>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>>find
>>>differences in amplifiers. It was called the null speaker test, again
>>>IIRC.
>>>
>>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>>so
>>>that the speaker would generate sound if there were any differences
>>>between
>>>the 2.
>>>
>>>Has anyone heard of this test, and if so, can you point me towards a
>>>place
>>>that describes the setup in detail?
>>
>> Cannot recall that but I do recall something similar that was advanced
>> by Hafler. Probably involves inverted polarity feeds to two amps with
>> the amp running off the two (+) terminals or a summing of the outputs.
>>
>> Kal
>
> I don't think you want to invert the feeds. Take two amps set at
> identical gain with a common feed, tie the grounds of each amp together
> (neg speaker terminal) and connect the positive output of one amp to the +
> speaker terminal, the positive of the other amp to the negative speaker
> terminal.
> The speaker will only "play" the difference of the amps.
>
> Seems to me if you invert the polarity to one amp you will always have a
> difference.
>
> I can't see how this method would be very revealing. Maybe if you took the
> difference signal and ran it through a 3rd amp to get some gain on it.
>
> ScottW
Gain is not the goal, finding a difference or not would be. That is the way
it was presented. IIRC I read about it in Electronics Now or maybe before
they changed the title and it was called Raido Electronics, probably written
by Larry Klein. I might still have the issue somewhere, I'll have to check.
If I find it I'll post what it is for sure.

Your set up sounds more correct than Roberts since inverting would certainly
present a difference.

[[email protected]]

"Clyde Slick" > wrote in message
...
>
> "AZ Nomad" > wrote in message
> ...
>> On Tue, 11 Oct 2005 22:53:15 GMT,
>> > wrote:
>>
>>
>>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>>find
>>>differences in amplifiers. It was called the null speaker test, again
>>>IIRC.
>>
>>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>>so
>>>that the speaker would generate sound if there were any differences
>>>between
>>>the 2.
>>
>>>Has anyone heard of this test, and if so, can you point me towards a
>>>place
>>>that describes the setup in detail?
>>
>>
>> P.S. learn how to crosspost. Starting multiple independant threads
>> in different newsgroups is assinine.
>
> Damned if you do, damned if you don't
My thoughts exactly.

[AZ Nomad]

On Wed, 12 Oct 2005 04:46:09 GMT, > wrote:



>"Clyde Slick" > wrote in message
...
>>
>> "AZ Nomad" > wrote in message
>> ...
>>> On Tue, 11 Oct 2005 22:53:15 GMT,
>>> > wrote:
>>>
>>>
>>>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>>>find
>>>>differences in amplifiers. It was called the null speaker test, again
>>>>IIRC.
>>>
>>>>Essentially it was a way to hook up 2 amplifiers and a single loudspeaker
>>>>so
>>>>that the speaker would generate sound if there were any differences
>>>>between
>>>>the 2.
>>>
>>>>Has anyone heard of this test, and if so, can you point me towards a
>>>>place
>>>>that describes the setup in detail?
>>>
>>>
>>> P.S. learn how to crosspost. Starting multiple independant threads
>>> in different newsgroups is assinine.
>>
>> Damned if you do, damned if you don't
>My thoughts exactly.


Damned if you post separate multiple identical messages and damned if you
crosspost to a thousand newsgroups. You're OK if you can gather three brain
cells to figure out how to crosspost a *single* message to a reasonable number
of newsgroups.

[Robert Morein]

" > wrote in message
ink.net...
>
> "Robert Morein" > wrote in message
> ...
> >
> > "Kalman Rubinson" > wrote in message
> > ...
> >> On Tue, 11 Oct 2005 22:53:15 GMT, "
> >> > wrote:
> >>
> >> >Years ago I read an article by Larry Klein IIRC, that describe a way
to
> > find
> >> >differences in amplifiers. It was called the null speaker test, again
> > IIRC.
> >> >
> >> >Essentially it was a way to hook up 2 amplifiers and a single
> >> >loudspeaker
> > so
> >> >that the speaker would generate sound if there were any differences
> > between
> >> >the 2.
> >> >
> >> >Has anyone heard of this test, and if so, can you point me towards a
> > place
> >> >that describes the setup in detail?
> >>
> >> Cannot recall that but I do recall something similar that was advanced
> >> by Hafler. Probably involves inverted polarity feeds to two amps with
> >> the amp running off the two (+) terminals or a summing of the outputs.
> >>
> >> Kal
> >
> > It consists of a bridge. One input to the bridge is the line-level input
> > to
> > the amplifier.
> > The second input to the bridge is the output of the power amplifier,
> > passed
> > through a voltage divider.
> > The bridge output consists of the difference signal between the input to
> > the
> > amplifier, and the output.
> >
> > Hafler claimed that the XL series of amplifiers exhibited a level of
> > distortion, when measured in this fashion, 20dB below other extant
> > amplifiers. The box was to be loaned by dealers so that XL owners could
> > make
> > a nulling adjustment in the field. The nulling adjustment is a capacitor
> > that adjusts an ultrasonic peak in the frequency response of the
> > amplifier,
> > thus providing an open loop compensation for load reactance at high
> > frequency. The XL design also claims an extremely low phase shift: 1/4
> > degree from 20Hz to 20kHz.
> >
> > Reviewers seemed to ignore this feature, preferring to test the
amplifier
> > only at the factory default setting. But if Hafler's claims are to be
> > believed, his method picked up, as an undifferentiated power, types of
> > distortion that were or are generally ignored by HM and IM measurements.
> > This is one reason I am skeptical of claims that amplifier distortion
has
> > been adequately quantified.
> >
> I don't know if this is the process I was trying to remember or not, but
the
> main reason I mention it is that it would be a very good way to find out
if
> there's any differnce between two amps. No sound, no difference, no
> difference, no upgrade. And you don't need an ABX box.
>
The Hafler box does not compare two amps. However, it does compare the
amplifier to the input, ie., to the mythical "straight wire with gain." It
will extract every change the amplifier has made to the input, and display
just the change.

One can hook a scope or spectrum analyzer up to the output of the box, and
see a representation of the distortion which is entirely different from the
description of TM and IM specs. Unlike those numbers, this test misses
nothing. Hafler claimed that his amplifiers were 20 dB cleaner than other
amplifiers, as of 1997, when the Excelinear design debuted. I use these
amplifiers extensively, and find them to be extremely revealing of low level
detail. They are a good match for metal domes. However, they are not nearly
as dynamic as amplifiers with othe topologies, such as the Transnova, or
bipolar amplifiers, because. This is because the design of conventional
MOSFET amplifiers prevents makes it impossible to drive them all the way
"on". For fabric domes, I recommend other amplifiers, such as your Acoustat.

Do not try to use this box with your Acoustat. The Hafler box cannot be used
with an amplifier that requires that both output terminals float from
ground.

Truely, Mikey, setting aside our jousting for a moment, I am sorry that you
are not receptive to these kinds of observations with respect to amps. You
are missing a world of perception, as well as fun, by sticking with your
notion that amplifiers sound the same. I have a very large personal
collection of power amps. Every time I go through the collection, I hear the
music in a different way.

[[email protected]]

"Robert Morein" > wrote in message
...
>
> " > wrote in message
> ink.net...
>>
>> "Robert Morein" > wrote in message
>> ...
>> >
>> > "Kalman Rubinson" > wrote in message
>> > ...
>> >> On Tue, 11 Oct 2005 22:53:15 GMT, "
>> >> > wrote:
>> >>
>> >> >Years ago I read an article by Larry Klein IIRC, that describe a way
> to
>> > find
>> >> >differences in amplifiers. It was called the null speaker test,
>> >> >again
>> > IIRC.
>> >> >
>> >> >Essentially it was a way to hook up 2 amplifiers and a single
>> >> >loudspeaker
>> > so
>> >> >that the speaker would generate sound if there were any differences
>> > between
>> >> >the 2.
>> >> >
>> >> >Has anyone heard of this test, and if so, can you point me towards a
>> > place
>> >> >that describes the setup in detail?
>> >>
>> >> Cannot recall that but I do recall something similar that was advanced
>> >> by Hafler. Probably involves inverted polarity feeds to two amps with
>> >> the amp running off the two (+) terminals or a summing of the outputs.
>> >>
>> >> Kal
>> >
>> > It consists of a bridge. One input to the bridge is the line-level
>> > input
>> > to
>> > the amplifier.
>> > The second input to the bridge is the output of the power amplifier,
>> > passed
>> > through a voltage divider.
>> > The bridge output consists of the difference signal between the input
>> > to
>> > the
>> > amplifier, and the output.
>> >
>> > Hafler claimed that the XL series of amplifiers exhibited a level of
>> > distortion, when measured in this fashion, 20dB below other extant
>> > amplifiers. The box was to be loaned by dealers so that XL owners could
>> > make
>> > a nulling adjustment in the field. The nulling adjustment is a
>> > capacitor
>> > that adjusts an ultrasonic peak in the frequency response of the
>> > amplifier,
>> > thus providing an open loop compensation for load reactance at high
>> > frequency. The XL design also claims an extremely low phase shift: 1/4
>> > degree from 20Hz to 20kHz.
>> >
>> > Reviewers seemed to ignore this feature, preferring to test the
> amplifier
>> > only at the factory default setting. But if Hafler's claims are to be
>> > believed, his method picked up, as an undifferentiated power, types of
>> > distortion that were or are generally ignored by HM and IM
>> > measurements.
>> > This is one reason I am skeptical of claims that amplifier distortion
> has
>> > been adequately quantified.
>> >
>> I don't know if this is the process I was trying to remember or not, but
> the
>> main reason I mention it is that it would be a very good way to find out
> if
>> there's any differnce between two amps. No sound, no difference, no
>> difference, no upgrade. And you don't need an ABX box.
>>
> The Hafler box does not compare two amps.

Then it is not the setup I was describing.

However, it does compare the
> amplifier to the input, ie., to the mythical "straight wire with gain." It
> will extract every change the amplifier has made to the input, and display
> just the change.
>
Display it? Do you mean that the speaker will prodcue sound?

> One can hook a scope or spectrum analyzer up to the output of the box, and
> see a representation of the distortion which is entirely different from
> the
> description of TM and IM specs. Unlike those numbers, this test misses
> nothing. Hafler claimed that his amplifiers were 20 dB cleaner than other
> amplifiers, as of 1997, when the Excelinear design debuted. I use these
> amplifiers extensively, and find them to be extremely revealing of low
> level
> detail.

I'm sure you believe that, but wothout a level matched, bias controlled
comparison, there is no way to know for sure.

They are a good match for metal domes.

In your opinion.

However, they are not nearly
> as dynamic as amplifiers with othe topologies, such as the Transnova, or
> bipolar amplifiers, because. This is because the design of conventional
> MOSFET amplifiers prevents makes it impossible to drive them all the way
> "on". For fabric domes, I recommend other amplifiers, such as your
> Acoustat.
>

I'm using a Focal t90ti02 inverted dome and it sounds fine.

> Do not try to use this box with your Acoustat. The Hafler box cannot be
> used
> with an amplifier that requires that both output terminals float from
> ground.
>
> Truely, Mikey, setting aside our jousting for a moment, I am sorry that
> you
> are not receptive to these kinds of observations with respect to amps.

I'm sorry you refuse to deal with the accepted standards for comparing audio
devices so that you can remove your bias from teh equation.

You
> are missing a world of perception, as well as fun, by sticking with your
> notion that amplifiers sound the same.

I've never said all amps sound the same and you know it, neither has anybody
who advocates ABX or any other form of DBT. We say repeatedly that amps
that measure closely enough. CLOSELY enough, that is within .1 db. More
than that and the ear can detect a difference, and then you can decide which
one you prefer.

I have a very large personal
> collection of power amps. Every time I go through the collection, I hear
> the
> music in a different way.
>
As would be expected from a non-bias controlled, non-level matched
comparison, as you should be well aware.

[Sander deWaal]

"Robert Morein" > said:

>The Hafler box does not compare two amps. However, it does compare the
>amplifier to the input, ie., to the mythical "straight wire with gain." It
>will extract every change the amplifier has made to the input, and display
>just the change.


An experiment I sometimes do with a just finished amp: connect a dummy
load consisting of a resistor and capacitor in parallel, and playing a
CD.

Then another amp of known reputation connected to the feedback signal,
reproduces the error signal that feedback loop nr. 1 creates.

Very revealing!

Of course, a double trace oscilliscope and a low distortion generator
will yield results that are better suited to produce on paper.

The Hafler box does the same IIRC, but with a HF compensation that has
to be nulled out or something (have to go through my stash of
magazines, I know it was described somewhere).

--

"Audio as a serious hobby is going down the tubes."
- Howard Ferstler, 25/4/2005

[Sander deWaal]

" > said:

>Your set up sounds more correct than Roberts since inverting would certainly
>present a difference.


Probable differences could also be due to certain phase shifts between
amplifiers, that wouldn't be a problem when using an amp stand-alone,
but might be audible in this setup.

Also, there are inverting and non-inverting amplifiers.
In case of comparing both sorts, one channel has to be inverted first.

--

"Audio as a serious hobby is going down the tubes."
- Howard Ferstler, 25/4/2005

[Ruud Broens]

"AZ Nomad" > wrote in message
...

: Damned if you post separate multiple identical messages and damned if you
: crosspost to a thousand newsgroups. You're OK if you can gather three brain
: cells to figure out how to crosspost a *single* message to a reasonable number
: of newsgroups.

Three wise brain cells from the east ? one man's reason is another man's folly
you will note
singles tend to cluster
:)

[Robert Morein]

"Sander deWaal" > wrote in message
...
> "Robert Morein" > said:
>
> >The Hafler box does not compare two amps. However, it does compare the
> >amplifier to the input, ie., to the mythical "straight wire with gain."
It
> >will extract every change the amplifier has made to the input, and
display
> >just the change.
>
>
> An experiment I sometimes do with a just finished amp: connect a dummy
> load consisting of a resistor and capacitor in parallel, and playing a
> CD.
>
> Then another amp of known reputation connected to the feedback signal,
> reproduces the error signal that feedback loop nr. 1 creates.
>
> Very revealing!
>
> Of course, a double trace oscilliscope and a low distortion generator
> will yield results that are better suited to produce on paper.
>
> The Hafler box does the same IIRC, but with a HF compensation that has
> to be nulled out or something (have to go through my stash of
> magazines, I know it was described somewhere).
>
You are thinking of the HF compensation on the Excelinear amp design, an
optimization adjustment.
The Hafler test box has no adjustment except for a level matching pot on the
voltage divider, so that the bridge can properly compute the difference.

OK, you have the feedback loop signal, but the actual correspondence of the
loop signal to the error in the output depends upon the loop gain. Do you
feel the observation gives as accurate an assessement of the error as a
nulled comparison of input to output?

[Robert Morein]

"Sander deWaal" > wrote in message
...
> " > said:
>
> >Your set up sounds more correct than Roberts since inverting would
certainly
> >present a difference.
>
>
> Probable differences could also be due to certain phase shifts between
> amplifiers, that wouldn't be a problem when using an amp stand-alone,
> but might be audible in this setup.
>
> Also, there are inverting and non-inverting amplifiers.
> In case of comparing both sorts, one channel has to be inverted first.
>
> --
>
> "Audio as a serious hobby is going down the tubes."
> - Howard Ferstler, 25/4/2005

Each amplifier has its' own characteristic, frequency dependent phase shift.
Ideally, a comparison method would allow one to time-delay the compared
signal before nulling, for minimum total power error. After all, an absolute
time shift, constant across the spectrum, has no impact on sound.

With two amplifiers, there are two phase shift curves to deal with. With the
Hafler box, there is one.

Sander, would you like me to scan the documentation and send it to you?

[Clyde Slick]

"AZ Nomad" > wrote in message
...
> On Wed, 12 Oct 2005 04:46:09 GMT,
> > wrote:
>
>
>
>>"Clyde Slick" > wrote in message
...
>>>
>>> "AZ Nomad" > wrote in message
>>> ...
>>>> On Tue, 11 Oct 2005 22:53:15 GMT,
>>>> > wrote:
>>>>
>>>>
>>>>>Years ago I read an article by Larry Klein IIRC, that describe a way to
>>>>>find
>>>>>differences in amplifiers. It was called the null speaker test, again
>>>>>IIRC.
>>>>
>>>>>Essentially it was a way to hook up 2 amplifiers and a single
>>>>>loudspeaker
>>>>>so
>>>>>that the speaker would generate sound if there were any differences
>>>>>between
>>>>>the 2.
>>>>
>>>>>Has anyone heard of this test, and if so, can you point me towards a
>>>>>place
>>>>>that describes the setup in detail?
>>>>
>>>>
>>>> P.S. learn how to crosspost. Starting multiple independant threads
>>>> in different newsgroups is assinine.
>>>
>>> Damned if you do, damned if you don't
>>My thoughts exactly.
>
>
> Damned if you post separate multiple identical messages and damned if you
> crosspost to a thousand newsgroups. You're OK if you can gather three
> brain
> cells to figure out how to crosspost a *single* message to a reasonable
> number
> of newsgroups.

many have been damned just for replying to a 2 newsgroup crosspost.