I understand that it increases as GNFB increases. But how would one determine how much is needed. I have very efficient older JBL's with the 15" D130 LF drivers in smallish vented cabinets. Would these need more or less DF? I haven't done the calcs, but I believe that I am running about 15db of GNFB.
Some confusion about definition?
What about Damping factor - Wikipedia?
More importantly, why do we have to worry about this? Does a poor damping factor affect the sound quality or the amplifier stability?
Regards ...
What about Damping factor - Wikipedia?
More importantly, why do we have to worry about this? Does a poor damping factor affect the sound quality or the amplifier stability?
Regards ...
The Wikipedea definition only adds in the speaker impedance. Since that is far from flat the actual damping factor is a curve. The spec — useless IMHO — is typically grabbed at 1kHz and 8 ohms.
Whether high or low output impedance (or in between) is good is directly related to the speake rthat is being used with the amplifier.
dave
Whether high or low output impedance (or in between) is good is directly related to the speake rthat is being used with the amplifier.
dave
If your driver has Re= 1 Ohm and your Damping factor is 15 then you can expect frequency response abnormalities. Why? Because the output impedance = 1/15 ~ 0.7 Ohms is on the order of the driver impedance. The amp and driver appear in series, and act like a voltage divider.
I would assert that even with an 8 Ohm driver, if your damping factor is only 15 then you are still disturbing the frequency response by a little bit, but over a wide range of frequency. Is this then something terrible that must be avoided? Not necessarily. You just need to know what will happen when the amplifer's output impedance is in series with the load (the loudspeaker or driver).
Remember, the driver presents a complex load to the amp. This load has an impedance magnitude that has one or more HUGE peaks unless the designer specifically added components to reduce this. Take a driver sans crossover. It might have Re=8 Ohms, but near the resonance frequency the impedance might be over 100 Ohms. At high frequency (depending on inductance) the impedance might slowly climb up to several tens of Ohms. The impedance is very "up and down". When you put this varying load in series with the amp's output impedance, and the amp's output impedance is not very small, what happens is that LESS POWER is delivered into the load when the load impedance is relatively LOW. Conversely, FULL POWER is delivered into the load when the load impedance is relatively HIGH. The difference between HIGH and LOW power in this case is the ratio of the amp's output impedance and the driver impedance as a function of frequency.
Most of the time this is no big deal. It is only with tube amps (generalizing here) that this becomes an issue. Why? Because they often have non-insignificant output impedance WRT the driver. This essentially causes an "EQ-ing" of the driver's output following the inverse of it's impedance curve. Some people actually like it, because it tends to accent low frequencies around resonance, and high frequencies where the impedance is rising. This just happens to "fix" some problems with some types of full range drivers. But in general it is not something good, because it is difficult to "control" from speaker to speaker, driver to driver, and amplifier designs will "sound" different if they have different output impedance. You are at the mercy of the impedance curve, or must implement lots of impedance compensation in your passive crossover.
Using a very low output impedance amplifier (e.g. damping factor > 100) tends to take any concerns of this type completely out of the equation. Typically this means a solid state output stage.
Last edited:
It definitely affects the sound.
Small DF will do a bass with little control as in tube amps.
Well, not really. The damping factor is defined as the dimensionless ratio of impedances, Z(load)/Z(source). Your definition (depending on precisely what is hidden in that twiddles sign!) seemingly has the damping factor being a dimensionful quantity with dimensions (ohms)^(-1).
But damping factors are a bit misleading anyway. One would get the impression from the definition that if the source were "ideal," i.e. Z(source) = 0, then the speaker would be "infinitely damped." (For example, that if the speaker cone were pushed inwards a bit, it would take an infinite time to return to its equilibrium position.) But that, of course, is not true, and in fact is very far from true. The resistance of the load should really be added into the denominator of the ratio defining the damping factor. Resistance in the load has just as much effect as the same amount of resistance in the source, as far as limiting the damping is concerned.
The situation is really quite a complicated one, when one looks properly at the dynamics of the speaker when it is driven by a source. A better measure of damping would be provided by something like the ratio
Z(speaker)/R(total),
where R(total) = R(speaker) + R(source). (Assuming, for simplicity, that the source (the amplifier) is purely resistive in its output impedance.) The question of how much of the speaker's impedance is accounted for by the pure resistance of the wire in the coil, versus how much is non-resistive reactance due to the dynamical motion of the speaker cone, is complicated. But quite a large proportion, depending on the frequency, is probably due to the resistance of the wire in the speaker coil, and that portion should be added into the denominator.
The twiddles is the symbol for "proportional to”.
Clearly DF=Z(load)/Z(source) implies DF ~ 1/Z(source) so exactly like i said.
dave
Damping factor is a very misleading spec and not nearly as useful as knowing the output impedance (Z(source), the speaker impedance (Z(load)) and how well damped the LF alignmnet of the speaker is.
There is an appropriate output impedance for any speaker. Most modern commercial loudspeaker is designed assumming a low output impedance (ie voltage source), but in the diy world we have a lot more choice.
Since a speaker is driver is a current controlled device (not a voltage controlled one) there are advantages to amplifiers with high output impedance (current amp).
What is appropriate in the amplifier is very much dependent on the speaker to be used with it. They should not be considered separately.
dave
There is an appropriate output impedance for any speaker. Most modern commercial loudspeaker is designed assumming a low output impedance (ie voltage source), but in the diy world we have a lot more choice.
Since a speaker is driver is a current controlled device (not a voltage controlled one) there are advantages to amplifiers with high output impedance (current amp).
What is appropriate in the amplifier is very much dependent on the speaker to be used with it. They should not be considered separately.
dave
This article may have the answers to your questions, noted the effect on frequency response of various damping. Once you have determined the critical damp factor of the speaker then you can adjust NFB for amp to lower or raise output impedance to match the critical damping so that the frequency response becomes flat. The output impedance of the amp and speaker damping factor need to be found out separately of course.
Last edited:
Yes, indeed, I agree with you about that. But also, as I was saying, the purely resistive component of the speaker's impedance should really be adding into the denominator, not the numerator, in the ratio defining the damping.
One could roughly model the speaker as having an ideal zero-resistance coil (e.g. made of superconducting wire), in series with a resistor R(load). The measure of how well damped it will be is then given essentially by Z/R(tot), with R(tot)=R(load)+R(source), and Z being the impedance (highly frequency dependent) of the superconducting coil in its dynamical environment as it moves, on the speaker cone, in the magnetic field of the speaker magnet.
I just measured the DC resistance of one of my Lowther DX3 speakers, and it is about 7 ohms. Looking at the impedance plots for the DX3 there is a huge peak, of order 100 ohms or more, at about 60 Hz. Outside the range 30 - 130 Hz, the impedance is about 10 ohms or less. So outside the 30 - 130 Hz range, the resistance of the coil accounts for most of the impedance of the speaker. This means the "true" damping factor over most of the audio range will never be more than about 2 at the most, even if the source has virtually zero output impedance. And whether the source has 0.01 ohms or 1 ohm output impedance hardly affects the damping at all.
Get a solid state amp (it has oodles of damping factor). Now add some resistance in series with the output, and listen. Do you like the sound? Try a different value. Eventually you will find a value that suits your speakers and your ears, and you can try to emulate that value in the next amp you build.
On high powered subs about as high a df as you can get.
(at least 300, better 500 - 2000+).
Use "thick wire" also.
Some will say it doesn`t matter but clearly they didn`t experiment themselves.
Except that you may be altering that amplifier's stability as you change the resistor, so what is heard may be a combination of the two. But I think this is a good way to see just how unimportant very high DF is.
Last edited:
Yes I suppose so, although the resistance you use will probably be only a fraction of the speaker coil resistance, so I suspect the stability change won't be very much in most circumstances.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.