Damping factor
Sportster Dave
William M. Johnson, Jr wrote:
>
>
> If I remember correctly, the damping factor is the ratio of the speaker's
> impedance to the amplifier's output impedance. An 8 ohm speaker with an
> amp with a .08 ohm output impedance would have a damping factor of 100.
> (Somebody correct me if I'm wrong, please!) Amps with a specified damping
> factor listed are usually rated against a standard
> speaker impedance which may vary from manufacturer to manufacturer and model
> to model.
>
> Oh, generally solid state amps have higher (better) damping factors,
Higher is NOT necessarily better; some horn or Hemholtz drivers actually
work better with amps with lower damping factors.
but another factor is
> how many output devices (tubes or transistors) are running in parallel in
> the output stage.
> Huge amps with a lot of output devices (like Krell or Levinson monoblocks)
> have very low output impedances and resulting large damping factors.
Patently not true!!! The transistors are CURRENT SOURCES and therefore
have HIGH impedance and therfore a very LOW damping factor!!! What is
done on SS amps is to use LOTS of NEGATIVE FEEDBACK!!!
With
> tube amps, the output transformer is more important. Designs using toroidal
> output transformers tend to have the highest damping factors in tube
> equipment, such as the Balanced Audio Technology VK-60.
Again, NOT TRUE!!! The toroidal OTs have a wider bandwidth, allowing the
use of...MORE NEGATIVE FEEDBACK!!!
>
> Hope this helps!
>
> Chris Johnson
> wjoh...@palmnet.net
>
> --
> ......................................................................
> Pursuant to U.S. code, title 47, Chapter 5, Subchapter II, Section 227
> Any and all unsolicited commercial E-mail sent to this address is
> subject to a fee of US $500.00. E-Mailing denotes acceptance of these
> terms. Consult <http://www.law.cornell.edu/uscode/47/227.html> for
> details.
Greg Leary
Can anyone tell me if a Damping Factor of >50 at 8 ohms is considered
high for a tube amplifier? My tube amps have this damping factor (i.e.,
greater then 50 at 8 ohms). How does this damping factor compare to say
solid state? What is the significance of damping factor? Thank you for
any help you can provide in answering this.
--
Sincerely,
Greg Leary (electronically signed)
William M. Johnson, Jr
Greg Leary wrote:
I can tell you what exactly the damping factor is, but I can't answer some
of the other
questions.
Damping factor is basically the effective output impedance of your amp as
measured
across the speaker terminals. Ideally, the impedance should be close to
zero as possible.
A simple demonstration of how the damping factor works:
Obtain a passive subwoofer in a sealed box. (No amp in the box, just input
terminals to
the speaker)
The box should be well sealed. With nothing hooked up to the woofer, when
you tap
gently but quickly on the cone, the woofer will make a tone which fades
away. Sort of a
'HMMmmm....' If you push gently on the cone with your hand, the cone will
move with a certain degree of freedom. Now take a short piece of bare wire
and connect the two terminals on
the box to each other and repeat the tap test. This time the woofer will
just give a dull thud with almost no ringing or tone at all. If you
gently push on the cone with your hand, it will resist you a lot more than
in the previous part of the test.
The push test will also work on a driver which is out of a cabinet entirely.
What is happening is the wire is allowing the electric force generated in
the voice coil in its motion through the magnetic field to go back into the
voice coil, generating a reverse magnetic field which acts against the
original movement.
This way the cone won't ring. It is desirable for the amplifier to act
like that shorting wire
because the cone will remain under better control. The amp doesn't do the
controlling of
excess cone movement, it merely allows the speaker to control itself.
If I remember correctly, the damping factor is the ratio of the speaker's
impedance to the amplifier's output impedance. An 8 ohm speaker with an
amp with a .08 ohm output impedance would have a damping factor of 100.
(Somebody correct me if I'm wrong, please!) Amps with a specified damping
factor listed are usually rated against a standard
speaker impedance which may vary from manufacturer to manufacturer and model
to model.
Oh, generally solid state amps have higher (better) damping factors, but
another factor is
how many output devices (tubes or transistors) are running in parallel in
the output stage.
Huge amps with a lot of output devices (like Krell or Levinson monoblocks)
have very low output impedances and resulting large damping factors. With
tube amps, the output transformer is more important. Designs using toroidal
output transformers tend to have the highest damping factors in tube
equipment, such as the Balanced Audio Technology VK-60.
Bob Bruhns
Greg, take Dave's corrections to Chris, and you pretty much
have it.
Damping factor is the ratio of load impedance to source
impedance. I consider a damping factor of >50 to be high for a
tube amp, and anything over about 5 to 10 is almost certainly
achieved by the use of negative feedback.
The effect of a high damping factor is as Chris said, it tends
to control dynamic speaker resonances. In most dynamic speakers
this is a good thing, but too much of this may be undesirable in
some speaker designs as Dave pointed out.
In most dynamic speaker systems, acoustic output tends to be
highest at the resonant frequencies, where the impedance tends to
be the highest as well. If the amplifier acts as a current
source (low damping factor/high source impedance), it will drive
more voltage into the higher impedances, accentuating the
resonant effect. But if the amplifier acts as a voltage source
(high damping factor/low source impedance), it will drive less
current into the higher impedances, which reduces the resonant
effect.
But if the speaker system requires a low damping factor to make
deliberate use of those resonances, then a very high damping
factor would hurt its response, instead of helping it.
Amplifier devices (tubes, transistors) -tend- to produce what
is called "current" output, (so-and-so many milliamperes of
output per volt, or per milliamp in some cases, of control
signal). This is the meaning of transconductance and beta.
There is also a resistive output property, which causes the
device characteristic to respond to output voltage in such a way
that it produces more output current into a low impedance load.
It acts like intrinsic negative feedback. This is what is meant
by plate resistance, it acts like resistance in series with a
zero impedance voltage output. Lower plate resistance is
equivalent to more intrinsic negative feedback. Triodes with low
amplification factor generally have the lowest plate resistance.
If output is taken from the cathodes, emitters or sources of
the amplifier devices, there is a strong voltage negative
feedback that causes the amplifier to have a low source
impedance. It tends to act as a "voltage source" producing an
output voltage of so-and-so many volts, without much regard to
the load impedance. Many solid state amplifiers use emitter or
source output, which intrinsically produces a low impedace output
(high damping factor), and they -also- use external negative
feedback. These amps usually have the highest damping factors
(>100).
When output is taken from plates, collectors or drains,
external application of strong negative feedback will have pretty
much the same effect, producing fairly high damping factors
(usually up to 50 or 100 or so).
My own opinion is that there is a limit to the effectiveness of
low damping factor, caused by the inherent resistance of the
voice coils, crossover networks (if used), and of the connecting
wire. At some point, even a large increase in damping factor
will have little effect on performance. But in the range of zero
to 20 or so, damping factor clearly makes a difference.
Bob Bruhns, WA3WDR, bbr...@erols.com
(No spam, please; you're wasting your time, I don't read it.)
Randy Nachtrieb
Hi Greg,
Damping Factor is the ratio of nominal load impedance(usually
referenced as 8 ohms) to the amplifiers effective output impedance.
It is important because it acts as a resistance in series with the
impedance of your loudspeaker. For optimum bass damping & flattest
frequency response, most speakers like to see as little additional
resistance(high DF) as possible.
The signifigance of DF largely depends upon the impedance
curve(varation of impedance with frequency) of your speaker. A
relatively uniform impedance curve means that the speaker's tonal
response is relatively insensitive to DF, as the voltage dropped across
the internal resistance of both the amplifier & speaker will likewise be
uniform.
Some speakers such as ribbons or Magneplanars have very flat
impedances, deviating perhaps 10% across the the audible range. Contrast
this, to one electostatic type, with an impedance of 2 ohms at the
highest frequencies vs 500 ohms in the bass region. Most cone type
speakers have characteristics that fall somewhere in between these two
extremes.
A DF of 50 is high for a standard xfmr coupled tube amp. For most
such amps, 15-20 is the norm. Transformerless amps such as OTL's & SS,
often have DF's exceeding 100 as the phase-shift of the output xfmr
usually limits the quanity of negative feedback which can be employed.
Many single-ended tube amps with little or no NFB have DF's as low as 2
or 3.
Generally speaking, a DF of 15 or more is sufficient, anything over
50 would probably be pointless. Some amps have DF's in the thousands or
even infinity, but the series resistance of the speaker cable will
generally limit them to an effectve DF of 100 or less.
An amplifiers DF is not necessarily constant either, often times
dropping off at the frequency extremes(not good). The amps with the
higher DF's are generally the ones most likely to do this, one way of
cheating on specifications.
It should however be remembered, that high DF is not necessarily a
universal virtue. There are some speakers( & systems) which were
designed around, or just happen to sound better at lower damping
factors. A high DF is however generally assumed by most speaker
designers, as it is their only practical universal reference point.
Regards,
Randy Nachtrieb
Enlightened Audio
New Art Loudspeakers
Daniel J. Marshall
Sportster Dave wrote:
>
> William M. Johnson, Jr wrote:
> >
> >
> > impedance to the amplifier's output impedance. An 8 ohm speaker with an
> > amp with a .08 ohm output impedance would have a damping factor of 100.
> > (Somebody correct me if I'm wrong, please!) Amps with a specified damping
> > factor listed are usually rated against a standard
> > speaker impedance which may vary from manufacturer to manufacturer and model
> > to model.
> >
> > Oh, generally solid state amps have higher (better) damping factors,
>
> work better with amps with lower damping factors.
>
> > how many output devices (tubes or transistors) are running in parallel in
> > the output stage.
> > Huge amps with a lot of output devices (like Krell or Levinson monoblocks)
> > have very low output impedances and resulting large damping factors.
>
> have HIGH impedance and therfore a very LOW damping factor!!!
Actually many, perhaps most, transistor amplifiers have complementary
emitter follower output stages, many times driven by a low impedance
emitter follower stage. This makes for a low output impedance, the
voltage gain being in a previous stage(s). Of course feedback further
reduces the output impedance. If I recall correctly from years ago, the
output impedance of an emitter follower is the base spreading resistance
plus the source impedance/beta. This can be quite low for a beefy
power transistor driven by a low impedance source. Even little
plastic small-signal transistor emitter followers can have an output
impedance on the order of 10 ohms or so.
Dan Marshall
What is
> done on SS amps is to use LOTS of NEGATIVE FEEDBACK!!!
>
> With
> > tube amps, the output transformer is more important. Designs using toroidal
> > output transformers tend to have the highest damping factors in tube
> > equipment, such as the Balanced Audio Technology VK-60.
>
> Again, NOT TRUE!!! The toroidal OTs have a wider bandwidth, allowing the
> use of...MORE NEGATIVE FEEDBACK!!!
> >
William M. Johnson, Jr
Sportster Dave wrote:
William M. Johnson, Jr wrote:
>
>
> If I remember correctly, the damping factor is the ratio of the speaker's
> impedance to the amplifier's output impedance. An 8 ohm speaker with an
> amp with a .08 ohm output impedance would have a damping factor of 100.
> (Somebody correct me if I'm wrong, please!) Amps with a specified damping
> factor listed are usually rated against a standard
> speaker impedance which may vary from manufacturer to manufacturer and model
> to model.
>
> Oh, generally solid state amps have higher (better) damping factors,
Higher is NOT necessarily better; some horn or Hemholtz drivers actually
work better with amps with lower damping factors.
but another factor is
> how many output devices (tubes or transistors) are running in parallel in
> the output stage.
> Huge amps with a lot of output devices (like Krell or Levinson monoblocks)
> have very low output impedances and resulting large damping factors.
Patently not true!!! The transistors are CURRENT SOURCES and therefore
have HIGH impedance and therfore a very LOW damping factor!!! What is
done on SS amps is to use LOTS of NEGATIVE FEEDBACK!!!
Hey, Dave, don't blow a gasket or anything here. I said that if I was mistaken on something, please correct me.
I appreciate the corrections, but could you back off on the excessive use of caps and exclamation points?
Besides, while I didn't specifically mention feedback (My omission,
sorry.) the basic concept of large numbers of
output devices and large damping factors following each other is true
in most cases. I merely forgot to finish the link between the cause
and the effect.
With
> tube amps, the output transformer is more important. Designs using toroidal
> output transformers tend to have the highest damping factors in tube
> equipment, such as the Balanced Audio Technology VK-60.Again, NOT TRUE!!! The toroidal OTs have a wider bandwidth, allowing the
use of...MORE NEGATIVE FEEDBACK!!!
Same comment as above.
--
......................................................................
Pursuant to U.S. code, title 47, Chapter 5, Subchapter II, Section
227
Any and all unsolicited commercial E-mail sent to this address is
subject to a fee of US $500.00. E-Mailing denotes acceptance of these
terms. Consult <http://www.law.cornell.edu/uscode/47/227.html> for
details.
"Gently, genteel..." --Cogsworth..Beauty and the Beast
Kirk R. Patton
CyberBlob wrote:
> Damping factor (audio amplifiers) is the measure of a circuits ability
> to overcome back EMF. Back EMF is power generated by the speaker in
> normal use.
This is, in fact, a classic defination, and the defination from where
the specification gets its name. But in the real world, given
voice-coil resistances, wire resistances, etc. the back EMF from the
speaker is really a very, very small amount of energy at best. Some
engineers (from highly respected audio companies) feel that the whole
back-EMF explanation is much more of an old-wives' tale than actual,
working theory.
But a high damping factor is still quite important. The more practical
application of this specification is as the output impedance of an
amplifier relates to the impedance curve of a speaker -- Randy Nachtrieb
wrote an excellent post regarding this.
Power bandwidth is something different altogether. Power bandwidth is
the ability of an amplifier's frequency-response and and
frequency-versus-distortion characteristics to remain constant at
different power levels. This is usually expressed as Bode plots
(frequency-response charts) at different power levels, THD+N versus
power at different frequencies, or THD+N versus frequency at different
power levels.
I'm not responding to be pissy, and your post was nothing to get
defensive about. No problems. Just trying to clarify a few things
where I can. . .
Regards,
Kirk Patton
CyberBlob
I'm probably off on the specifics :) so I'll just say, here's the
general idea. I won't ask for corrections, because I'll get them anyway,
even if they aren't correct. If someone gets real pissy about it, I will
dig out the books and post the definitions. :P
Damping factor (audio amplifiers) is the measure of a circuits ability
to overcome back EMF. Back EMF is power generated by the speaker in
normal use. The higher the damping factor, the more the amplifier will
overcome the power generated by the speaker. A lower damping factor
amplifier will have a harder time dealing with the current the speaker
generates, you may have a peak at the speakers resonant frequency. In a
circuit, damping is used to prevent the circuit from oscillating, unless
of course it's an oscillator :o hahaha. A higher damping factor may
indicate a lower output impedance on an amplifier. A damping factor of
greater than 50 @ 8 ohms is fine (IMHO) for any audio amplifier, as my
dusty ears remember it all :o anything above a 10 or so sounds generally
good to me. I don't remember the origins of the actual specification,
but I believe it usually pops up during power bandwidth measurements.
Well I am already blabbing :o so I'll stop now :) I hope this helps
For those who will pick this to death, go ahead, I didn't get specific
enough to be useful :) and I will dig up the books for the next round.
Heh :p
In article <34CAB...@bellsouth.net>, greg...@bellsouth.net sez...
Greg Leary
Thanks to all who responded to my question regarding damping factor.
Your answers have given me the necessary insight in trying to determine
if my amps can be used to "control" certain dynamic speakers. As of
this posting I appreciate William M. Johnson, Jr., Sportster Dave,
Daniel J. Marshall, Bob Bruhns and CyberBlob for responding to my
inquiry. Also, thanks to anyone else who responds. Have a great day!
Greg Leary
Thanks to you and all others for answering my post.
Arny Krüger
>Greg Leary wrote:
>
> Can anyone tell me if a Damping Factor of >50 at 8 ohms is considered
> high for a tube amplifier?
That seems pretty good for a tubed amp. Its really quite good for a tubed
amp.
>My tube amps have this damping factor (i.e., greater then 50 at 8 ohms).
At what frequencies?
>How does this damping factor compare to say solid state?
At low frequencies, 50 might be pretty poor for a SS amp.
One thing to consider is that damping factor is not properly stated as a
single number. For just about all amps, damping factor depends on frequency.
The usual rule for tubed amps is that damping is better at the midrange,
than at the low or high ends. For SS amps, damping usually is best at low
frequencies and drops as frequency increases.
> What is the significance of damping factor?
Low damping factor goes hand-in-hand with unpredictable (really, harder to
predict) frequency response with loudspeaker loads. Show me a power amp with
low damping factor (under say under 40) and I'll show you a power amp that
makes different speakers sound more different from each other. Show me an
amp that has a damping factor greater than say, 100 over most of the audible
range, and I'll show you an amp that has a better than average chance of
being pretty transparent.
Show me a power amp with a damping factor under 10 and I'll show you an
equalizer with no knobs. ;-)
Steve Jones
Arny Krüger wrote:
>
> Show me a power amp with a damping factor under 10 and I'll show you an
> equalizer with no knobs. ;-)
With speakers not properly designed for use with an amp with a high
output impedance, I'd agree.
-Steve Jones
Sheldon D. Stokes
Fisher Z-matic, and the Heathkit Designs like the UA-1 with current
feedback. They have a damping factor of "1"
As a note, the Heath UA-1's (also known as UA-2, or AA-61) sounds
absolutely awful on the Quad ESL's when in the unity damping factor mode.
So lets' see this knobless EQ. :)
Sheldon
--
Remove SPAM_BE_GONE. from my address to reply to me.
rjs...@calcube.com
Arny Krüger wrote:>
> Show me a power amp with a damping factor under 10 and I'll show you an
> equalizer with no knobs. ;-)
Do buttons count?;)
--
Robert J. Salvi, Ambiance Acoustics
California Cube Loudspeaker System
http://www.calcube.com
San Diego, CA USA
619-485-7514
Michael Wong
>>How does this damping factor compare to say solid state?
>
>At low frequencies, 50 might be pretty poor for a SS amp.
>
>One thing to consider is that damping factor is not properly stated as a
>single number. For just about all amps, damping factor depends on frequency.
>The usual rule for tubed amps is that damping is better at the midrange,
>than at the low or high ends. For SS amps, damping usually is best at low
>frequencies and drops as frequency increases.
My Adcom manual says that my amp's damping factor is greater than 1000
from 20Hz to 20kHz. Does this mean that it is actually much higher in
the midrange but they don't want to bother stating it that way?
>> What is the significance of damping factor?
>Low damping factor goes hand-in-hand with unpredictable (really, harder to
>predict) frequency response with loudspeaker loads. Show me a power amp with
>low damping factor (under say under 40) and I'll show you a power amp that
>makes different speakers sound more different from each other. Show me an
>amp that has a damping factor greater than say, 100 over most of the audible
>range, and I'll show you an amp that has a better than average chance of
>being pretty transparent.
I've heard this from other sources as well- is that very high damping
factor of >1000 just useless grist for the advertising mill, then?
>Show me a power amp with a damping factor under 10 and I'll show you an
>equalizer with no knobs. ;-)
I think Radio Shack can oblige you there :)
Stewart Pinkerton
nospam...@ebtech.net (Michael Wong) writes:
>>>How does this damping factor compare to say solid state?
>>
>>At low frequencies, 50 might be pretty poor for a SS amp.
>>
>>One thing to consider is that damping factor is not properly stated as a
>>single number. For just about all amps, damping factor depends on frequency.
>>The usual rule for tubed amps is that damping is better at the midrange,
>>than at the low or high ends. For SS amps, damping usually is best at low
>>frequencies and drops as frequency increases.
>
>My Adcom manual says that my amp's damping factor is greater than 1000
>from 20Hz to 20kHz. Does this mean that it is actually much higher in
>the midrange but they don't want to bother stating it that way?
Yes, but anything over fifty is pointless anyway.......
>>> What is the significance of damping factor?
>
>>Low damping factor goes hand-in-hand with unpredictable (really, harder to
>>predict) frequency response with loudspeaker loads. Show me a power amp with
>>low damping factor (under say under 40) and I'll show you a power amp that
>>makes different speakers sound more different from each other. Show me an
>>amp that has a damping factor greater than say, 100 over most of the audible
>>range, and I'll show you an amp that has a better than average chance of
>>being pretty transparent.
Show me an amp that has a damping factor above 100 at the speaker
terminals, and I'll show you an amp connected by a couple of feet of
welding cable..................
>I've heard this from other sources as well- is that very high damping
>factor of >1000 just useless grist for the advertising mill, then?
>
>>Show me a power amp with a damping factor under 10 and I'll show you an
>>equalizer with no knobs. ;-)
Yes indeed, 20-50 is a good useable range, although more than 50 does
no harm, so long as stability and HF distortion do not suffer as a
result.
--
Stewart Pinkerton | Music is art, audio is engineering
ASP Consulting |
(44) 1509 880112 |
Greg Leary
Arny Krüger wrote in part:
>
> >Greg Leary wrote:
> >
> > Can anyone tell me if a Damping Factor of >50 at 8 ohms is considered
> > high for a tube amplifier?
>
> That seems pretty good for a tubed amp. Its really quite good for a tubed
> amp.
>
<snip, snip,cut, cut>
Both Arny Kruger and Stewart Pinkerton (and maybe others) "wondered or
questioned" the number 50 as a damping factor for a tube amp. Let me
clarify: I have LA Audio P-66 mono blocks (no, not Los Angeles amps, but
rather, the LA stands for Lennart Andersen -- his original company was
LA Audio but now is Audio Design, Denmark). The original brochure,
written in English but produced in Denmark, listed the damping factor as
I originally stated: greater then 50 at 8 ohms. Well, to make a long
story somewhat shorter, I later emailed Lennart to verify the damping
factor. His response was that the damping factor of my P-66's in 8 ohm
load is approximately 12. My apologies to all if I have caused any
confusion or bewilderment. However, I do greatly appreciate all the
explanations regarding damping factor.
Chris Johnson
To clarify, our (Sonic Frontiers) POWER Series of amplifers have a
damping factor in this region.....>30 for the POWER-1 (55 watt stereo)
and >50 for both the Power-2 (110 watt stereo) and Power-3 (220 watt
mono). This is accomplished through a variety of
techniques...including the careful implementation of feedback - both
positive and negative...the total amount amplied being comparable to
other commercially available designs.
As has been pointed out by several contributors to this thread, this
measure of output regulation describes the amplifer's ability to
prevent the loudspeaker's impedance curve affecting the frequency
response of the amp. In our experience, a damping factor of under 10
to 12 will, in most cases, cause the amplifier to act as a partial
tone control...and a haphazard one at best. A damping factor beyond
50 is essentially superfluous since the frequency response abberations
it will allow are below the threshold of human hearing (and certainly
below the driver tolerances of most loudspeakers).
It is important to note that damping factor does not describe the
amplifier's ability to "control" the woofer...this is a myth given the
cumulative affect of the downstream impedance characteristics of the
speaker cable/cross-over/voice coil combination.
Regards,
Chris Johnson,
President - Sonic Frontiers, Inc.
On Sun, 01 Feb 1998 02:25:59 GMT, Greg Leary <greg...@bellsouth.net>
wrote:
Anonymous
Chris Johnson wrote:
>As has been pointed out by several contributors to this thread, this
>measure of output regulation describes the amplifer's ability to
>prevent the loudspeaker's impedance curve affecting the frequency
>response of the amp. In our experience, a damping factor of under 10
>to 12 will, in most cases, cause the amplifier to act as a partial
>tone control...and a haphazard one at best.
>
>It is important to note that damping factor does not describe the
>amplifier's ability to "control" the woofer...this is a myth given the
>cumulative affect of the downstream impedance characteristics of the
>speaker cable/cross-over/voice coil combination.
Chris,
These two statements are contradictory. A high damping factor reduces the
tone-control effect of an amplifier/speaker combination *precisely by*
controlling the drivers -- by making them do what the input signal tells
them to do instead of what their momentum would otherwise make them do.
Any amplifier/cable/crossover/voice coil combination that cannot "control"
the drivers cannot reduce the tone-control effect either.
>A damping factor beyond 50 is essentially superfluous since the frequency
>response abberations it will allow are below the threshold of human hearing
>(and certainly below the driver tolerances of most loudspeakers).
Not necessarily. It has been shown many times that human hearing can
detect very small EQ variations. But the *frequency response* variations
are not the most troublesome sonic artifacts of poor radiator control
anyway (they could always be fixed with EQ). Uncontrolled radiators allow
the sound to spread out in time, which destroys imaging cues and detail.
This time-domain distortion cannot be easily fixed.
I have shown to my satisfaction that damping factors up to 150 or so
(wideband) at the speaker terminals improve the sound of most audiophile
speakers. To deliver this one needs mono amplifiers connected to the
speakers with a foot or less of speaker wire.
Henry Pasternack
Anonymous (Use-Author-Address-Header@[127.1]) wrote:
: These two statements are contradictory. A high damping factor reduces the
: tone-control effect of an amplifier/speaker combination *precisely by*
: controlling the drivers -- by making them do what the input signal tells
: them to do instead of what their momentum would otherwise make them do.
: Any amplifier/cable/crossover/voice coil combination that cannot "control"
: the drivers cannot reduce the tone-control effect either.
The motion of a loudspeaker driver is never "controlled" by the drive
signal in the sense that, say, the input to a servo system controls the
position of an actuator. In fact, is is precisely the "decoupling" (one
might say) of the driver motion from the drive signal that allows a
dynamic driver to have flat frequency response.
At low frequencies, cone acceleration is low and the position of the
cone is "stiffness controlled", depending on the force generated by the
motor balancing the restoring force due to the spider and surround. The
SPL output depends on acoustic radiation resistance, which is increasing
with frequency as long as the acoustic wavelength is large compared to
the cone diameter. Therefore, at low frequencies the driver output is
rising, at a second-order rate of 12dB/octave, I believe.
At some low frequency (relative to the driver passband) there will
be a mechanical resonance due to the moving mass of the driver and the
combined compliance of the suspension and the air in the enclosure.
Above this frequency, the cone motion is now mass-controlled and the
cone excursion falls with increasing frequency given constant drive
voltage. Whether or not there is a peak at resonance depends on the
mechanical and electrical damping. Best bass response does not occur
with maximum or "infinite" damping. Infinite damping would translate
to infinite friction in the driver and would result in no output at
all.
In the midband, since the frequency of operation is above the primary
system resonance, which is also second-order, the transfer function of
cone motion versus drive frequency has a constantly-varying phase shift
that approaches 180 degrees. The cone is not under the "control" of the
amplifier. Rather, the amplifier pushes and pulls, and the cone somewhat
lazily responds, as though it is a weight on the end of an imaginary
rubber band which the amplifier grasps and moves up and down at the other
end.
What saves the day is that the poles due to the driver mechanical
resonance cancel the zeros due to the rising acoustical radiation
resistance. The system, then, has flat acoustical output and phase
over a range of several octaves.
At some higher frequency, still, the cone diameter becomes equal to
the acoustical wavelength and the radiation resistance ceases to rise.
The mass-controlled response characteristic reasserts itself and the
system output begins to decline again at 12 db/octave. This analysis
assumes the driver is on a large baffle and does not take into account
the effect of baffle directivity.
Since the DC voice coil resistance of a typical 8 Ohm driver is on
the order of 5-6 Ohms, it's clear that beyond an amplifier damping
factor of ten or so, the changes in system response due to driver
"control" are minimal. This will manifest itself primarily as small
changes in bass -3dB point and peaking/damping. However, since the
electrical impedance of the speaker system typically varies all over
the place, there is still plenty of opportunity for a highish output
impedance to lead to audible response variations.
-Henry
--
ATTENTION! Reply to h...@nortel.ca (hen...@nortel.ca won't work).
Kirk R. Patton
Anonymous wrote:
>
> Chris Johnson wrote:
> >It is important to note that damping factor does not describe the
> >amplifier's ability to "control" the woofer...this is a myth given the
> >cumulative affect of the downstream impedance characteristics of the
> >speaker cable/cross-over/voice coil combination.
>
It occurred to me that this myth wouldn't be too hard to dispell, ahem,
test, with actual amplifiers and speaker cables . . .
Build a cabinet where two woofers share a common, sealed air space. One
woofer is connected to a drive amplifier -- the drive being the signal
used to test the damping factor -- let's say a plain ol' sine wave. The
second woofer has an accelerometer mounted on the cone, like a Velodyne
servo sub. The voice coil of the accelerometer-equipped driver is
attached to whatever piece of wire, crossover network, amplifier, etc.
that you wish to test the "driver control ability" of. The signal from
the accelerometer could then be compared between the
"amplifier-dampened" condition and an open-circuit condition. Computer
analysis would be handy here, and not too tough to do . . . could also
allow for testing with real music.
Anybody care to try something like this? I don't have the time . . .
Regards,
Kirk Patton
Anonymous
Henry wrote:
>The motion of a loudspeaker driver is never "controlled" by the drive
>signal in the sense that, say, the input to a servo system controls the
>position of an actuator.
I do not believe this is contrary to my claim, if that was intended. I did
not say that loudspeakers are controlled by drive signals "in the sense
that...the input to a servo system controls the position of an actuator."
>Best bass response does not occur with maximum or "infinite" damping.
>Infinite damping would translate to infinite friction in the driver and
>would result in no output at all.
I'm not sure what you mean here. Infinite electrical damping corresponds to
the inability of externally applied forces to move the cone when there is
no input signal (which is a desirable condition). But it does not follow
that there would be no output with an input signal.
"Best" bass response depends on many factors. It indeed occurs with a high
amplifier DF if the speaker designer assumed voltage-source drive (as most do).
Good damping (however achieved -- electrically or mechanically) suppresses
resonances and the frequency-domain and time-domain distortions that
resonances introduce. Rarely can designers arrange sufficient mechanical
damping at all frequencies, so they usually rely on electrical damping from
the amplifier as well. Due to unavoidable losses in the system (e.g.,
voice-coil resistance), the electrical damping available to the designer
is never perfect even with a perfect voltage source, and consequently all
real speakers exhibit unwanted resonances. The question is how much
reduction of them we can get through low amplifier output impedance.
Kirk wrote:
[To test the degree of control amplifiers exert over speakers]:
>Build a cabinet where two woofers share a common, sealed air space. One
>woofer is connected to a drive amplifier
>
>The second woofer has an accelerometer mounted on the cone, like a Velodyne
>servo sub. The voice coil of the accelerometer-equipped driver is
>attached to whatever piece of wire, crossover network, amplifier, etc.
>that you wish to test the "driver control ability" of. The signal from
>the accelerometer could then be compared between the
>"amplifier-dampened" condition and an open-circuit condition.
If comparison to no damping is what you want, just thump a woofer cone with
your finger with the speaker's input terminals open and then shorted. The
difference is not subtle with most speakers. Amplifiers with reasonable
damping factors approximate the shorted condition, not the open condition.
Replace the short with the lumped networks representing each amplifier's
output impedance (or with the amplifiers themselves) and the differences
*between the various networks* are more subtle (thus less amenable to the
simple thump test).
The discusssion here on RAT has been directed toward how high a damping
factor needs to be (i.e., how low the output impedance of the amplifier
needs to be) before further DF increase is inaudible. Several people have
said than this occurs at around DF = 10 (RO = .8 ohms). For speakers driven
through substantial lengths of cable, fuses, speaker-level crossovers,
etc., this may be a useful rule of thumb (though IMO even so it is too
low). For speakers with active crossovers driven through short cables it is
(again, IMO) way too low.
If one can tell the difference between two amplifiers that differ only
in their output impedances, IMO the lower-impedance amplifier is a priori
better. Why? one might ask -- given the roller-coaster frequency response
of real speakers, what justifies calling a response difference of tenths of
a dB better a priori? It is not the response difference per se that makes
the one better. Response differences could always be adjusted by
equalization. The lower output impedance amplifier also dissipates the
energy from inertial cone motion sooner, resulting in better time-domain
performance (which cannot easily be corrected). It is this better
time-domain performance that justifies calling the one amplifier better.
[Remember, the premise is that there is an audible difference -- if
there is no audible difference neither amplifier is better.] FWIW, the best
way I have found to experiment with this is to start with an amplifier that
has very high (>200) DF from DC to 50+ kC and to simulate higher output
impedances with lumped series networks.
One last point on damping: Damping debates (including this one) are often
carried on with a tacit assumption that damping is important only at the
system resonant frequency of the woofer. This is an erroneous assumption.
Many speakers rely on electrical damping throughout the spectrum. The old
Advent loudspeaker is a classic example: it has a wicked HF resonance
that makes it sound horribly harsh and aggressive on amplifiers with poor
HF damping (sad to say, many high end SS amplifiers to this day suffer from
low damping at high frequencies). Used with an amplifier that holds its DF
above 20 in the top octave, the Advent smooths right out and gives the
restrained "New England sound" it was known for.
Kirk R. Patton
>
Enjoyed the post, and also the others on SRPP. My main point is that
back-EMF-shorting ability of the amplifier is only one, and definately
not the most important, aspect of the complex relationshp between
amplifier output impedances and speaker systems . . . other models,
especially the time-domain aspect you mention, are much more important
in optimizing amplifier DF.
> Henry wrote:
> >Best bass response does not occur with maximum or "infinite" damping.
> >Infinite damping would translate to infinite friction in the driver and
> >would result in no output at all.
I belive that what Henry is referring to here is the total mechanical
and electrical damping of the loudspeaker cone with regards to applied
mechanical forces, inertia and input signal included; in the same way
"infinate damping" in a vehicle suspension system is, well, a rigid
frame, i.e. no motion.
> If comparison to no damping is what you want, just thump a woofer cone with
> your finger with the speaker's input terminals open and then shorted. The
> difference is not subtle with most speakers. Amplifiers with reasonable
> damping factors approximate the shorted condition, not the open condition.
The "thonk" test is pretty flawed; drumming on a speaker cone produces
resonances and harmonics that are way outside the speaker's operating
frequency, and won't work when multiple in-phase drivers share a common
airspace . . . this is a good way to sell someone a servo-controlled
subwoofer but not an accurate model of the effects of DF on back EMF.
But honestly, I haven't walked around drumming on woofers in quite
awhile, and it would be easy for me to try, so I'll do it.
> The discusssion here on RAT has been directed toward how high a damping
> factor needs to be (i.e., how low the output impedance of the amplifier
> needs to be) before further DF increase is inaudible. Several people have
> said than this occurs at around DF = 10 (RO = .8 ohms). For speakers driven
> through substantial lengths of cable, fuses, speaker-level crossovers,
> etc., this may be a useful rule of thumb (though IMO even so it is too
> low). For speakers with active crossovers driven through short cables it is
> (again, IMO) way too low.
Most people's arguements about how high DF needs to be directly
correlate to how high the DF is in their amplifier of choice . . . ref.
Chris Johnson's comments to the specs on Sonic Frontiers amps . . .
(although beautiful, well-engineered equipment IMO). I feel that the
most important aspect in the implementation of feedback to an amplifer
is the time-domain performance, the achievable DF is often more related
to the phase margin of the circuit in question rather than any specific
numerical goal. Yes, damping factor should be as high as possible, and
linear over the audio spectrum and beyond, given that the basic
stability issues of the amplifier are properly dealt with.
> If one can tell the difference between two amplifiers that differ only
> in their output impedances, IMO the lower-impedance amplifier is a priori
> better. Why? one might ask -- given the roller-coaster frequency response
> of real speakers, what justifies calling a response difference of tenths of
> a dB better a priori? It is not the response difference per se that makes
> the one better. Response differences could always be adjusted by
> equalization. The lower output impedance amplifier also dissipates the
> energy from inertial cone motion sooner, resulting in better time-domain
> performance (which cannot easily be corrected). It is this better
> time-domain performance that justifies calling the one amplifier better.
> [Remember, the premise is that there is an audible difference -- if
> there is no audible difference neither amplifier is better.] FWIW, the best
> way I have found to experiment with this is to start with an amplifier that
> has very high (>200) DF from DC to 50+ kC and to simulate higher output
> impedances with lumped series networks.
Well said. The people most concerned with ultrahigh damping factors
seem to be the live-sound-reinforcement amplifier manufacturers and
their marketeers; but this definately a moot point -- given the amount
of speaker cable snaking up to a flown array or to a large sub matrix,
plus the effect of frequently corrosion-and-contaminant-ridden
Speakons. I have also heard of no data regarding the correlation
between high-DF amplifiers and excersion-related woofer failures in such
applications -- as one might expect if the high-DF amplifiers were in
fact "damping" woofer motion.
> One last point on damping: Damping debates (including this one) are often
> carried on with a tacit assumption that damping is important only at the
> system resonant frequency of the woofer. This is an erroneous assumption.
Absolutely! But the woofer cone has the most inertia, and would
definately generate the bulk of the back-EMF produced by the speaker
system. So as far as the "motion-damping" arguements regarding DF, a
woofer is the most appropriate model.
> Many speakers rely on electrical damping throughout the spectrum. The old
> Advent loudspeaker is a classic example: it has a wicked HF resonance
> that makes it sound horribly harsh and aggressive on amplifiers with poor
> HF damping (sad to say, many high end SS amplifiers to this day suffer from
> low damping at high frequencies). Used with an amplifier that holds its DF
> above 20 in the top octave, the Advent smooths right out and gives the
> restrained "New England sound" it was known for.
Oh yeah. Anybody who uses electrostats (myself included) can attest to
this problem. Incidently, have you measured DF-versus-frequency of
various amplifiers? If so, I would be greatly interested in any
specifics or generalizations on this data . . .
Best Regards,
Kirk Patton
Randy Nachtrieb
Hi Kirk,
The "woofer tapping" test is really not very indicative of DF; this
is primarily because it not only excites the fundamental resonance, but
the diaphrams breakup modes as well.
Unless the diaphram is acting as a rigid piston, DF will be
ineffective at damping resonances. This is evidenced by the fact that
you will almost never see signifigant deviations from a drivers nominal
impedance except at fundamental resonance, or the gradual rise created
by the VC inductance. If signifigant damping of diaphram resonances were
taking place, drive unit impedance curves would often be quite erratic,
such is not the case.
The Advent situation is most likely related to impedance
interaction with the tweeters Fs and or xover network, or the VC
inductance; rather than having anything to do with the damping of a HF
resonance.
Regards,
Randy Nachtrieb
Anonymous
Kirk writes:
>The "thonk" test is pretty flawed; drumming on a speaker cone produces
>resonances and harmonics that are way outside the speaker's operating
>frequency, and won't work when multiple in-phase drivers share a common
>airspace . . .
>
>But honestly, I haven't walked around drumming on woofers in quite
>awhile, and it would be easy for me to try, so I'll do it.
It's no definitive test, for sure. But the difference between no
electrical damping and reasonable electrical damping is so great that even
this crude test reveals it starkly. By manipulating the cone in several
different ways (push, thump, etc.) you can get a little more information.
Try to cause voice-coil motion, not just cone flex (thump the cone near the
voice coil, in the direction of cone motion rather than perpendicular to
the cone surface).
BTW, I recommend against trying the thump test on your electrostatics. ;-)
>[H]ave you measured DF-versus-frequency of various amplifiers? If so, I
>would be greatly interested in any specifics or generalizations on this
>data
Yes, I've measured lots of them.
Generalizations:
Many, many high end SS amps have DFs down in the low double-digits at 20kc,
some even worse. (Amplifiers with Zobel output networks are usually
particularly poor at high frequencies, both because the networks themselves
have substantial impedance and because amplifiers that need them are poor
performers at high frequencies to begin with.) A few have DFs of 50-200
out to at least 20kc, and a very few remain >50 beyond 50kc. I consider SS
state of the art (best DF without incurring other, more audible defects) to
be 150-200 out to 50kc and >50 at 100kc. It helps to apportion local and
global NFB to give a more or less constant DF-vs-frequency characteristic,
rather than the common gazillion at 5 cycles declining at 6dB/8ve (this
tends to make the closed-loop output impedance resistive instead of
reactive).
High-end tube amplifiers: Most P-P NFB designs are in the 10-20 range in
the midband. P-P low-NFB designs are in the 2-10 range in the midband.
I'm shocked at how soon many of them fall off at both ends of the spectrum
(not just in DF, but in power response and distortion as well). It
sometimes helps to use a lower impedance tap (I almost always end up
listening to commercial tube amps on the lowest available tap). Using
pirated Mac, Citation, Acro, and Dyna output transformers one can readily
get DFs >30 from 10 to 40k cycles or better (with correspondingly good
power response and distortion). Unfortunately, many high end manufacturers
use transformers that aren't as good as these, and too often their
designers pursue some "better idea" with such vigor that they ignore other
defects to the detriment of the sound of the design.
From what I have seen of high end tube state of the art, the knowledgeable
hobbyist can do lots better for a small fraction of the price. Having said
that, I'll add that there are dozens of little tube high end companies, and
I haven't come close to seeing all their products.
Single-ended no-NFB designs: The ones I have seen have totally inadequate
damping. The only way to get reasonable damping without NFB is to
sacrifice lots of output power (a 211, 845, or 8003 putting out 15W, or
four 6336 sections putting out 5W, could have reasonable damping, but the
transformer would still have to carry the DC that the same tubes operating
to full capacity would require).
Randy Nachtrieb
Another note on DF; aside from response variations due to impedance
interaction & the damping of the fundamental resonances, DF is
unimportant.
The thought that a high DF will improve the time domain performance
of a speaker, through braking of the inertial mass of the
diaphram(except at resonance), is naive. The inertia is unimportant,
because the diaphram(or any radiating body for that matter) only
radiates when accelerating. Thus, except at resonance, DF will have no
effect upon the time domain character of the propagated waveform. Series
level dropping resistors are used in speakers all the time, with no ill
effects, provided the impedance interaction is accounted for.
Randy Nachtrieb
New Art Loudspeakers
Anonymous
>Another note on DF; aside from response variations due to impedance
>interaction & the damping of the fundamental resonances, DF is
>unimportant.
>The thought that a high DF will improve the time domain performance
>of a speaker, through braking of the inertial mass of the
>diaphram(except at resonance), is naive. The inertia is unimportant,
>because the diaphram(or any radiating body for that matter) only
>radiates when accelerating. Thus, except at resonance, DF will have no
>effect upon the time domain character of the propagated waveform.
Hi Randy,
I have not claimed much more than what you say here and in your other post.
You are correct that damping is not much help for diaphragm flex modes, and
that damping will help piston resonant modes. But the fundamental
(mass-compliance) resonance of the woofer is not the only piston resonance
in a speaker. There is a fundamental resonance for each driver, and
midrange drivers and tweeters are often worked down to them (though in a
perfect world they wouldn't be). In addition, there are usually several
compliances at work in each driver which may or may not all work on the
whole radiator mass (i.e., the driver is modeled not as one mass on a
spring, but as several masses coupled to each other by various springs and
resistances and to "ground" by several springs). The venting of midrange
drivers and tweeters and the various "innovative" acoustical loadings of
woofers all produce these additional resonances, some of which are
controlled to a greater or lesser degree by electrical damping.
To whatever degree any resonances are controlled by electrical damping (and
in a perfect world it would only be the fundamental resonance of the
woofer, or maybe not even that if we could get it low enough), they
contribute less to the time domain errors of the speaker. Resonances
produce time-domain errors (in addition to frequency domain errors), and
suppressing resonances suppresses the time domain errors related to the
resonances. Damping does not affect non-resonant time-domain errors
(crossover group delay, etc.), but I have not claimed that it does.
Henry Pasternack
Anonymous (Use-Author-Address-Header@[127.1]) wrote:
: To whatever degree any resonances are controlled by electrical damping (and
: in a perfect world it would only be the fundamental resonance of the
: woofer, or maybe not even that if we could get it low enough), they
: contribute less to the time domain errors of the speaker. Resonances
: produce time-domain errors (in addition to frequency domain errors), and
: suppressing resonances suppresses the time domain errors related to the
: resonances. Damping does not affect non-resonant time-domain errors
: (crossover group delay, etc.), but I have not claimed that it does.
It wasn't my intention to contradict your original posting so much
as to reinforce the notion that "infinite" damping, the hypothetical
"vice grip" of the woofer by an amplifier, does not give best bass
response. I also wanted to make the point again that driver motion
does not precisely mirror the input signal amplitude, even in a properly
operating ideal loudspeaker. Obviously, you are not confused on this
point, but many audiophiles (and audiophile pundits) are.
With respect to damping at frequencies other than the primary driver
resonance, I find it more helpful to view these as causing ripples in
the speaker's impedance curve and to treat the resulting effects in the
electrical domain. Since the speaker is an electro-mechanical device,
any mechanical resonance will, after all, be reflected in the driving
point impedance. Anyway, many of these resonances are associated
with non-linear mechanisms (like cone breakup), or are not easily
controlled by lowering the source impedance (like enclosure resonance).
There seems to be a lot of audiophile confusion in this area. They
accept that a crossover network, containing inductors and capacitors,
presents a varying electrical load and requires low-impedance drive for
flat frequency response. But they view the storage and subsequent
release of electromechanical energy as a separate phenomenon. Back EMF
seems like some kind of mysterious fluid that insinuates itself into
feedback loops and magically interferes with the amplifier operation.
Only by sopping up all of this pernicious fluid, it seems, can proper
operation of the speaker and amplifier be achieved.
I would like to emphasize that electrical and mechanical resonance
are analogs of one another, and from the point of view of the signal at
the speaker terminals, driver mass and compliance are no no different
from any other electronic reactance. The speaker is supposed to move
when you push on the cone, and a speaker that does not do so (assuming
no servo) will not produce any bass, or, for that matter, any sound at
all.
Anyway.
Kurt Strain
Henry Pasternack (hen...@nortel.ca) wrote:
: It wasn't my intention to contradict your original posting so much
: "vice grip" of the woofer by an amplifier, does not give best bass
: response. I also wanted to make the point again that driver motion
: does not precisely mirror the input signal amplitude, even in a properly
: operating ideal loudspeaker. Obviously, you are not confused on this
: point, but many audiophiles (and audiophile pundits) are.
You should have heard what I heard last week. I helped someone set up a
pair of JMLabs Utopia speakers ($30K/pr!) driven by an Airtight KT-88 based
push-pull amp. This has by no means the best damping factor on earth,
obviously. But the bass was by far the most controlled and articulate
I have ever heard anywhere. Not to mention the detail everywhere else
was just pouring out in spades. Absolutely an ear-opening experience.
And it wasn't even broken in yet, and it wasn't even spiked in yet, and
it wasn't even optimally set up in position yet. But it still blew away
all others in bass performance than anything else I have heard, by a
huge margin. They don't need Krell amps to do it, they do it on their
own by good drivers, crossovers, and cabinetry. And the impedance does
dip down to a minimum of 3.5 ohms.
But I have not heard the Audio Artistry Beethoven's. And
I'm told those might set the standard for bass performance, if not these.
But the Utopias did it all, and seamlessly, with unbelievable detail in
imaging with it. The front end wasn't shabby, though, with VPI TNT, 12"
JMW arm, and the latest Koetsu Rosewood Signature.
Kurt
Henry Pasternack
Kurt Strain (kst...@sr.hp.com) wrote:
: But the bass was by far the most controlled and articulate I have ever
: heard anywhere.
Doesn't surprise me.
: But I have not heard the Audio Artistry Beethoven's. And I'm told those
: might set the standard for bass performance, if not these.
I got Siegfried's (Linkwitz, not Duraybito) AES article on the compact
dipole which was, presumably, a precursor to his work at Audio Artistry.
I'm really interested in trying dipole bass. I don't suppose you can
catch him in the cafeteria and wrestle some trade secrets from him, can
you?
Randy Nachtrieb
Hi Kurt,
Actually, bottom end performance is one of the few things which is
easy to get right on a loudspeaker, if you know what you're doing.
Randy Nachtrieb
Kurt Strain
Henry Pasternack (hen...@nortel.ca) wrote:
: : But I have not heard the Audio Artistry Beethoven's. And I'm told those
: : might set the standard for bass performance, if not these.
: I got Siegfried's (Linkwitz, not Duraybito) AES article on the compact
: dipole which was, presumably, a precursor to his work at Audio Artistry.
: I'm really interested in trying dipole bass. I don't suppose you can
: catch him in the cafeteria and wrestle some trade secrets from him, can
: you?
He's retired from HP as of last year. Maybe he sold enough speakers.
Kurt
Randy Nachtrieb
Henry,
Dipole bass ? Surely you jest.
Regards,
Randy
SBench
>
> Dipole bass ? Surely you jest.
>
>
this was going to be the end all to get rid of the Alewife (spelling)
problem in Lake Michigan, then they came up with the Coho
Salmon instead.
:-)
Steve
Henry Pasternack
Randy Nachtrieb (gat...@webtv.net) wrote:
: Dipole bass ? Surely you jest.
I understand your skepticism. And yet Linkwitz designed a series
of dipole loudspeakers which are marketed under the "Audio Artistry"
label and which have been extremely well-reviewed. The advantage of
the dipole bass is the (relative) lack of enclosure resonances and
the fact that something like 1/3 less bass energy ends up in the room
reverberant field.
The disadvantages are obvious. The question is how Linkwitz made
it practical.
Randy Nachtrieb
Hi Henry,
Upon reflection(no pun), I could see how a dipole could have some
interesting qualities(in terms of efficiency etc.), if made sufficiently
large.
I appreciate the nature of your interest, IMHO room interaction is
the most under-engineered aspect of loudspeaker design in general.
Though, dipoles don't really do that much to improve things in that
regard(null is limited to right angles), do also consider, a decrease in
reverberant energy of 1/3 is relatively insignifigant in the scheme of
things.
Lack of enclosure resonances is again an often cited advantage as
well; but consider, if the woofer box is limited to only the lowest
frequencies(lets say <100hz), it is not particularly difficult through
proper bracing & the useage of relatively stiff, lightweight materials,
to keep the mechanical resonances beyond the range of operation.
By the same token, there will be no standing waves within the
enclosure provided its longest internal dimension does not exceed 1/4
wavelength.
Another quality commonly attributed to dipoles(usually panel
types), is superior LF transient response. In reality it is inferior to
a simple sealed box, as the dipole has a roll-off rate of 18db/octave vs
12db/octave for the sealed enclosure.
Anyway, enough spilling my guts for now.
Regards,
Randy
T. L.
Hi Randy,-
Randy Nachtrieb <gat...@webtv.net> skrev i artikkelen
<6cf5r2$63j$1...@newsd-151.iap.bryant.webtv.net>...
> Another quality commonly attributed to dipoles(usually panel
> types), is superior LF transient response. In reality it is inferior to
> a simple sealed box, as the dipole has a roll-off rate of 18db/octave vs
> 12db/octave for the sealed enclosure.
> Anyway, enough spilling my guts for now.
> Regards,
> Randy
>
Ehm,-A little simplistic, wouldn't you say..
Could you elaborate ?
T.Lien
>
Daniel J. Marshall
Randy Nachtrieb wrote:
>
> Hi Henry,
>
> Upon reflection(no pun), I could see how a dipole could have some
> interesting qualities(in terms of efficiency etc.), if made sufficiently
> large.
> I appreciate the nature of your interest, IMHO room interaction is
> the most under-engineered aspect of loudspeaker design in general.
> Though, dipoles don't really do that much to improve things in that
> regard(null is limited to right angles), do also consider, a decrease in
> reverberant energy of 1/3 is relatively insignifigant in the scheme of
> things.
> Lack of enclosure resonances is again an often cited advantage as
> well; but consider, if the woofer box is limited to only the lowest
> frequencies(lets say <100hz), it is not particularly difficult through
> proper bracing & the useage of relatively stiff, lightweight materials,
> to keep the mechanical resonances beyond the range of operation.
> By the same token, there will be no standing waves within the
> enclosure provided its longest internal dimension does not exceed 1/4
> wavelength.
> types), is superior LF transient response. In reality it is inferior to
> a simple sealed box, as the dipole has a roll-off rate of 18db/octave vs
> 12db/octave for the sealed enclosure.
> Anyway, enough spilling my guts for now.
>
> Regards,
>
> Randy
I have always heard that the rolloff rate of a dipole is 6 dB/octave. I
am speaking of rolloff from cancellation losses where the speaker fs
substantially lower than the break point caused by cancellation. Were
the fs and rolloff from backwave cancellation to occur at the same
frequency, then the rolloff would be 18 dB/octave, as you suggest, but I
would not think one would design to this criteria, rather it would
seem appropriate to make fs quite low, substantially below the
cancellation break frequency (which is related to baffle dimensions),
and use +6dB slope compensation at the cancellation break point to
extend the f3 down as far as practical, perhaps to fs and the break it
out then let the system roll of at 18 dB/octave at a fairly low
frequency.
Dan Marshall
Randy Nachtrieb
Hi Dan,
Thanks for the re-iteration :). If the baffle is sufficiently small
to require much EQ, you had better have pretty long voice-coils, or alot
of woofers(in one sense or another). But I do agree, starting off with a
very low Fs offers considerable advantages.
Regards,
Randy Nachtrieb
Bob Fitzgerald
Daniel J. Marshall wrote in message
<6cg9hc$6...@bgtnsc03.worldnet.att.net>...
>I have always heard that the rolloff rate of a dipole is 6 dB/octave. I
>am speaking of rolloff from cancellation losses where the speaker fs
>substantially lower than the break point caused by cancellation. Were
>the fs and rolloff from backwave cancellation to occur at the same
>frequency, then the rolloff would be 18 dB/octave, as you suggest, but I
>would not think one would design to this criteria, rather it would
>seem appropriate to make fs quite low, substantially below the
>cancellation break frequency (which is related to baffle dimensions),
>and use +6dB slope compensation at the cancellation break point to
>extend the f3 down as far as practical, perhaps to fs and the break it
>out then let the system roll of at 18 dB/octave at a fairly low
>frequency.
>
>Dan Marshall
An easy way to extend the "size" of the baffle, thus lower the transition
frequency from
2Pi to 4 Pi radiation, is to place the bass module against the side wall and
floor.
I have heard the Beethovens, and they are VERY impressive. With the 4 12"
woofers in each of the two subs, there is a lot of air being pushed around.
But no "tubbiness", "boom" or other room related problems. Not recommended
for being driven by tube amps, though.
Bob
Bill Sheppard
The only real info gained is whether the voice coil is rubbing. We used
to check used speakers (unmounted) by thumping the back of the magnet,
not the cone. A clear 'thonk', free of any scratchiness, indicates no
V.C. rub.
Bill (o.c.)
Jamie
Bill Sheppard <old...@webtv.net> wrote in article
<6cpi73$smv$1...@newsd-143.iap.bryant.webtv.net>...
That sort of works, except that cones with a large excursion range
sometimes get in trouble at the very extremes of their excursion ranges. I
generally use a cup to apply even pressure around the center and push the
cone in and out several times to check for rub.....
--
Jamie Campbell
jcc at wwa dot com
Tubes are absolutely the WORST sound reproducers in the world, except for
everything else.