??? Damping Factor ???
Philip Witham
: Is there any general agreement about the effect of power-amplifiers
: damping factor? (or output impedance?) How much it depends on
: speakers?
I don't know what the verdict is in the midrange and treble, but the
Damping Factor has a known effect on the bass. A lower figure represents a
higher resistance "in series" with the woofer coil's resistance. This
resistance lowers the woofer's electrical damping, and raises the Q of
the woofer-box system. Just how that sounds would depend on the
speakers, but it's a fair bet that the bass will be tighter sounding, and
more controlled with a higher DF amp. It will reach a point of
diminishing returns, though, that is, 1000 is probably not noticeably
better than 300.
- Philip
pet...@ento.csiro.au
> Is there any general agreement about the effect of power-amplifiers
> damping factor? (or output impedance?) How much it depends on
> speakers?
bass but that is not necessarily due to damping factors. Indeed I could
imagine that an amp with a lower damping factor might sound better with
some speakers. This gets to a question. [Dick Pierce might like to come
in here. If I recall correctly he mentioned a strong view on this isn the
past but not discussed it at length] My (incomplete) understanding of the
theory is that when one designs a speaker one considers the various
electrical and mechanical impedances in series with the driver and designs
the box according to the degree of damping desired. This is a set of
compromises between bass extension, flatness of response, and ringing on
transients. In addition to the voice coil impedance (say 3-16 ohms) are
the series resistance of any series inductors in the crossover, the
*anticipated* series resistance of the speaker wire and the *anticipated*
output impedance of the amplifier. The "damping factor" is just the ratio
of the speaker's impedance to the amplifiers output impedance. The latter
is usually very low due to the use of feedback. Thus I believe it is
argued that as long as the output impedance of the amp is lower than the
speaker by a factor 10 or more then it has little effect on the tuning of
the speaker.
SO:
1) Have I got it right that the theory suggests that damping factor above a
threshold value has little effect of the tuning of a speaker? (I presume
that other factors can effect the quality of bass from an amp)
2) What do speaker designers assume in practice? Is a zero impedance amp
and speaker cable assumed because this close to the usual situation?
Perhaps a higher impedance is assumed leading to greater damping than
intended if one uses fat cable and a very high damping factor amp and
causes an over damped response?
Peter Campbell (pet...@ento.csiro.au)
C/O Div. Entomology, C.S.I.R.O.
Box 1700 Canberra A.C.T. 2601 Ph.61-6-2464158 (w), 61-6-2516213 (h)
Australia 61-6-2464173 (fax)
Richard D Pierce
> Is there any general agreement about the effect of power-amplifiers
> damping factor? (or output impedance?) How much it depends on
> speakers?
More on this in a subsequent post...
> I recently replaced my old Yamaha amplifier (damping factor 50) with
> Onkyo A-809 (damping factor 150), and the difference was very sig-
> nificant. I can't come up with any other reasonable explanation
> since i don't believe in "magic".
With no information forthcoming about how you were able to control other
variables (gain, memory, and so on), I see no data here that suggests
that whathever differences you heard can be attributable unambiguously to
damping factor. How, for example, did you insure that the gain of the two
were matched to within 0.1 dB so that the small gain differences, which
are known to trigger the impression of other qualitative differences, did
not exist in you comparison?
You don't need magic to explain differences, you only need experimental
errors such as gain. Once you've eliminated those, and you still can
reliably here a difference, then it's time to go exploring for a cause.
--
| Dick Pierce |
| Loudspeaker and Software Consulting |
| 17 Sartelle Street Pepperell, MA 01463 |
| (508) 433-9183 (Voice and FAX) |
Richard D Pierce
This has been debunked enough times that it ought to have stuck, but, I
guess not.
Philip, you make a very specific, testable assertion here, let's see if
your assertion withstands some simple tests.
Let's take your assertion a lower damping factor raises the Q of the
woofer-box system, and that it is, indeed, a "fair bet that the bass will
be tighter sounding" with a higher damping factor.
Let's do so by taking the best of your numbers, a DF of 1000, and compare
it to the effect of an amplifier with a DF of, say, 30. Here we have a 20
to 1 range in damping factor, and 50 is a factor of 10 below what you
claim is the threshold of insignificant improvement. Let's further assume
that we are driving, as you suggest, a closed box loudspeaker system
whose system Q (Qtc) is precisely tuned to 0.707107, a perfect 2nd order
Butterworth alignment whose response is maximally flat. Let's also assume
the mechanical losses for the system result in a system mechanical Q (Qmc)
of about 3, and the resulting electrical Q is about 0.925. The woofer has
a typical DC resistance of 6.5 ohms.
According to Small and others, the electrical part of the Q is dependent
upon source resistance in the following fashion:
Re + Rs
Qec' = Qec ---------
Re
where Qec is the 0-source system electrical Q, Re is the DC resistance of
the driver, and Rs is the source resistance presented by the amplifier.
A damping factor 0f 30 implies a source resistance of 1/30 that of the
"nominal" load impedance, usually 8 ohms, implying a source resistance
of, therefore 0.27 ohms. Then, the system's new electrical Qec' will be:
6.5 + 0.27
Qec' = 0.925 * ------------
6.5
Qec' will be equal to 0.964.
Now, combine that with the mechanical system Qmc and we get a new system
Q of:
0.964 * 3.0
Qtc = ------------
0.964 + 3.0
The new Qtc' will be equal to 0.729.
Indeed, the woofer-box Qtc is higher. By all of 3%. What effect does this
change have on the response? Is it, indeed, a "fair bet" that the result
is a looser sounding bass? Well, let's see.
The response peak at resonance is dependent upon system Qtc in the
following fashion:
4
Qtc 1/2
GjwMAX = [-----------]
2
Qtc - 0.25
Substituting a Qtc of 0.729, we get a bump at resonance that is
<drum roll, please>
A whopping 0.015 dB high.
I don't think that it is a "fair bet" that the difference between a
system with such a peak, and one that is dead flat, constitutes the
difference between a system that is "tighter sounding and more
controlled and one that is not.
A further fatal flaw in your assertion is that this 3% difference in Qtc
is well within the manufacturing tolerances of high quality woofers AND
is within the range of Qtc variations that will occur because of changes
in temperature causing material and acoustic changes.
No, it is an excellent bet that the attention paid to damping factor and
the effect on system Q is a complete misdirection of effort.
Let's turn the question around: what minimal damping factor is required
to assure that we will not influence the system Qtc enough to keep a
maximally flat system from developing, say, a 0.1 dB peak at resonance?
That one is easy to solve, just reverse the process above. The answer is
that the minimum damping factor required to meet the criteria in the
system described above is:
<'nuther drum roll, please>
results from a source resistance of about 0.75 ohms giving a system Qtc'
of about 0.768, meaning, a damping factor into a nominal 8 ohm load of
<cymbal crash>
11, eleven, XI, what have you. Remember, that's for a 0.1 dB peak. For a
0.5 dB maximum peak, you'll need a minimum damping factor of all of 4.
There may or may not be an audible effect of source resistance. However,
it's quite demonstrable that the old boogieman of "damping factor"
mucking about with the system Qtc simply isn't a likely mechanism for such
audible effects.
Petri Kekkonen
damping factor? (or output impedance?) How much it depends on
speakers?
I recently replaced my old Yamaha amplifier (damping factor 50) with
Onkyo A-809 (damping factor 150), and the difference was very sig-
nificant. I can't come up with any other reasonable explanation
since i don't believe in "magic".
- Petri Kekkonen - kekk...@cc.oulu.fi - rock...@phoenix.oulu.fi -
- Department of Theoretical Physics, University of Oulu, Finland -
_/_/_/_/ _/_/_/_/ _/_/_/_/ _/ _/ _/_/_/_/ _/_/_/_/ _/_/_/_/ _/_/_/
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
_/_/_/_/ _/ _/ _/ _/_/ _/_/_/ _/ _/ _/_/_/_/ _/ _/
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
_/ _/ _/_/_/_/ _/_/_/_/ _/ _/ _/ _/_/_/_/ _/ _/ _/_/_/
Carl.W.M...@att.com
>>Petri Kekkonen (kekk...@sun2.oulu.fi) wrote:
>>: Is there any general agreement about the effect of power-amplifiers
>>: damping factor? (or output impedance?) How much it depends on
>>: speakers?
>>
>>I don't know what the verdict is in the midrange and treble, but the
>>Damping Factor has a known effect on the bass. A lower figure represents a
>>higher resistance "in series" with the woofer coil's resistance. This
>>resistance lowers the woofer's electrical damping, and raises the Q of
>>the woofer-box system. Just how that sounds would depend on the
>>speakers, but it's a fair bet that the bass will be tighter sounding, and
>>more controlled with a higher DF amp. It will reach a point of
>>diminishing returns, though, that is, 1000 is probably not noticeably
>>better than 300.
FOr what it's worth, January's Sterophile has some measured freq.
response curves of various amplifiers into a selection of speakers.
Freq. response was measured at the speaker terminals.
Most of the amplifiers were very flat into all the speakers, except
perhaps for some hf rolloff.
One tube amp, however, showed wild response variations which followed
the speakers impedance curves.
Ah ha! I thought, but then I saw that the output impedance of the
particular amp was 3 ohms for a damping factor of a whopping 2.6.
And who says all amps sound alike :-).
Carl
Carl Muhlhausen
att!taz!ledzep
(908)-949-3402