Tube Amp for Martin Logan Speakers

Unsound ... I found that same point confusing too, especially in light of ML's recommendation to use an amp that increases power as speaker impedance drops. That sounds like a Voltage Paradigm/SS amp. But our other "enthusiastic" techie members wore me out. Glad you picked up on that point too.

Unsound and Bruce, as you may have noted TJN's article that Unsound referenced addresses how impedance interactions affect frequency response at the speaker's INPUT terminals. In itself that says nothing about what the speaker's acoustic output will be like in response to those inputs.

I have no knowledge as to whether optimal frequency response of the acoustic output of the Martin-Logan Aerius would result from a flat frequency response at its input, or from a frequency response at its input that is non-flat in some manner. But see the comments in my previous post regarding the Quad ESL-57, and take a look at its impedance curve that I referenced.

Best regards,
-- Al

So, to recap, what did we learn:

- output impedance of a 'box' is *defined* as the change in output voltage divided by the current change that caused the voltage variation;
- nfb makes the output voltage change smaller for the same current change, so *by definition* lowers the output impedance.

We also saw that nfb does not change the 'internal' output impedance of an amp - nfb operates strictly outside the amp proper. Also, we saw that (to a first order) this change in output impedance due to nfb does not change the amps output capability. Lowering the output impedance towards zero does not all of a sudden produce an amp that can deliver gobs of power in very small loads. The as-designed amp capability does not change.

A corollary we did not discuss but should be quite obvious: since damping factor is *defined* as load impedance divided by output impedance, output impedance drop due to nfb consequently increases damping factor in the same measure.

This was a good exchange!

Yes, fascinating thread, even for a non-techie like me.

If I may ask a corollary question: is there a relatively simple rule of thumb to determine whether a speaker might prefer a Power Paradigm or Voltage Paradigm amp?

:) Yes- damping factor is different from "output impedance" in that it gives you an idea of the servo performance of the circuit, it it has any.

Unsound, what you see from that graph is the woofer driver in a box, combined with the impedance curve of the ESL. The Aerius is a hybrid speaker. Interestingly, we also see that as ML speakers go, this one is less severe, dropping to a fairly manageable 2 ohms at 20KHz.

If I can draw your attention to the letter that is appended to this article, you can see from it why the ZERO works as well as it does, effectively lowering the output impedance of a zero feedback tube amplifier into the range where it might conform with the parameters laid out in the letter.

The response plots initially do seem to suggest that the speaker is Voltage Paradigm, since such amplifiers seem capable of producing flat frequency response. I feel its important to point out that the amplifiers in question all show a peak in response as I forecast, which seems to start up at about 7KHz, and increasing with frequency as the impedance of the speaker continues to drop.

Since the ear can hear a change in a spectrum of frequencies much better that it can with a single frequency, we now see that there is a correlation between the measurements and subjective listening experience wherein the amplifiers are causing the speaker to have brightness on the top end.

As I have mentioned before, this is a common complaint when combining ESLs and transistors.

The whole point of the Voltage Paradigm when it was proposed way back in the late 1950s and into the 1960s was that it would eliminate tonal coloration due to frequency response errors. What we see from these measurements is that isn't happening.

Now the Power Paradigm does not make any such guarantee. Instead, it seeks to keep the kinds of distortion to which the ear is most sensitive to a minimum. This is does because the ear translates such distortions to tonality, the idea being that there is a tipping point in the brain where the tonality of distortion can be more pronounced than actual frequency response errors. If this pans out correctly, it may well mean that the result will seem more linear, even when the frequency response does not seem to be quite as flat.

IOW, this whole thing has to do with how we perceive sound as opposed to how we measure it. Its a classic argument, and as I pointed out in my article about the Voltage and Power Paradigms (http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php) this argument has been around ever since the Voltage Paradigm was introduced.