Can an amp overdamp a speaker?

Gs5556, I agree with you that the primary low-frequency effect of amplifier output impedance (or damping factor) is on driver electrical Q. And the lowest possible value of Qe is not necessarily ideal.

Let me give an example of a speaker that would be suited for a low damping factor (high output impedance) amplifier. Suppose you have a very low-Q fullrange single-driver speaker whose frequency response (measured on a high damping factor amplifier) shows a gentle rolloff starting in the upper bass region, with the low bass being pretty weak. This speaker will sound thin and gutless without help in the bass region.

Okay, let's look at what happens when this speaker is driven by a tube amp with a very low damping factor, let's say a damping factor of 1 (output impedance of 8 ohms). This would be like doubling the driver's electrical Q. So in effect we would be getting a "free lunch" - about 3 dB or so of bass boost. This will probably make the speaker's tonal balance acceptable without having to add a subwoofer.

To give a real-world example, I build a speaker that is -3 dB at 35 Hz and -6 dB at 32 Hz with a solid state amp. With a low-damping-factor tube amp, this same speaker is -3 dB at 30 Hz and -6 dB at 24 Hz (this assumes appropriate re-tuning of the port.) As you can see, there is a significant "free lunch" here as well: the -6 dB point (which often predicts the actual in-room bass extension) moves about half an octave lower with the low damping factor amp. And, the response is actually smoother (less ripple) with the low damping factor amp.

I just gave away one of my secrets. Since it's over here under "Amps Preamps" instead of under "Speakers", maybe no one will notice.

Duke

Duke, I agree that using speaker equations can show that the lower cutoff frequency is a function of Qe. For example, if you assume a speaker design with a Qt of 0.707 with a driver that has a voice coil resistance of 5 ohms, a Qm of 3, 8 ohm impedance, and a Fs of 40 hz, then Qe at a DF of 1000 or greater will be 0.925 and give a lower cutoff of 40 hz. If you go down to a DF of 10, Qe is 15% higher and the lower cutoff drops 3 dB to 36 hz. But that's not a free lunch...

What you have to take into account is that the SPL ultimately depends on pressure generated by the driver. The factors affecting pressure are the driver magnet force factor, piston area, voice coil resistance and mechanical compliance. The SPL using these factors is inversely proportional to voice coil resistance. So decreasing the DF has the same effect as increasing the coil resistance so any -dB frequency drop by increasing Qt is offset by the reduction in SPL which is a wash. Otherwise, you could simply place a variable resistor in line with the driver which you can then adjust Qt to play with the frequency response in your room, independent of amplifier power.

Gs5556, the net effect of low damping factor on SPL is not the same as having a resistor in series. The modification of Qe is the same, but because there is in fact no resistor in series that is soaking up power there is no corresponding loss in SPL.

Once again, a real-world example: The low damping factor OTL amplifier I use delivers about twice the wattage into my speaker's high impedance in the bass region as it does into the 8-ohm load they present across the rest of the spectrum. That extra wattage in the bass region is indeed a free lunch.

Just for the record, note that the wattage output of a high damping factor amplifier is also modulated by the speaker's impedance curve, but in the opposite direction.

Duke

Duke, is an amplifier's ability to damp a speaker a function of several things including damping factor, amplifier power, the resistance of the speaker and resistance in the cable?

All other things being equal, wouldn't an amplifier with 500 watts into 4-ohms and a damping power of 1000, for instance, have a much higher damping ability than a 50 watt amplifier into the same load with the same damping factor?

Dave

Dcstep, I don't know how much of a difference greater amplifier power makes. In my experience if the amplifier and speaker work well together, it doesn't really matter what the amplifier's damping factor is.

I used to think that a speaker would ring like a bell if the amp didn't have a high damping factor, but I no longer believe that. The voice coil is powered all the way through its cycle. In a properly designed speaker, the voice coil never has to depend on the amplifier's low output impedance to "put on the brakes" and stop unintended oscillation.

I have owned two amplifiers (Berning ZH-270 and Wolcott Presence P220M) that had variable damping factor controls, which imho was a useful feature for pairing up with different speakers. With neither amp did I ever encounter a speaker that sounded best with the damping factor at the highest setting.

I tend to like zero-global-feedback tube amps, which inherently tend to have a very low damping factor. This does limit speaker choice somewhat, but once I put on my speaker designer hat I see opportunity there instead of limitation.

Duke