Tube Amp for Martin Logan Speakers

Bruce, thank you for your thoughtful response. It's late for me now, I'll try to find time in the near future for a more deserving answer. In the mean time, you might want to reread the chronology of the posts above. I think you'll understand why I ended that post to Atmasphere with a :-).

Unsound, I don't see the comments you found doing anything but supporting my position, and also this:

http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

The bit about Jim Theil using drivers in parallel to get more current out of the amp at low frequencies is a classic Voltage Paradigm move. The problem is it also increases the distortion of the amplifier.

As the other quotes you found also state (paraphrasing) - if sound **quality** is your goal then there is no argument for 4 ohms. If sound **pressure** is the goal then there is a weak argument for 4 ohms if you have a solid state amplifier that supports that operation.

Now in the case of Jim Theil, by using two woofers in parallel, and assuming an amplifier that can produce constant voltage into most loads, the simple result will be that it makes twice as much power into the woofer array than it would if there was only one woofer. Its also likely that the woofers chosen may well be 3 db less efficient than the midrange and tweeter, if by any chance the latter are 8 ohms.

Now in the case of a Martin Logan ESL panel, the panel impedance is not the result of the behavior of a driver in a box. It is the result of a capacitor, whose range of impedance is set by a matching transformer. As a result, its not to your advantage to see an increase in power at low impedances- the panel is as efficient at 10KHz as it is at 1KHz, even though the 10KHz impedance is much lower.

It is for this reason that it is perfectly reasonable for one to use a set of ZEROs on the speaker, even though there is a matching transformer in the speaker, and perhaps another one in the tube amplifier employed. IOW, the 4 ohm tap on many tube amplifiers may not be enough to allow the amp to deal with the 0.5 ohm impedance at 20KHz that is common with many ML ESLs. Heck, a lot of transistor amps have trouble with that too- especially if a speaker cable is being used between the amp and speaker.

Atmasphere, My disagreement was and continues to be with your earlier post(s):
"Now it is a simple fact that there is no good argument for lower impedance (4 Ohms or less) in high end audio; that is to say if **sound quality** is your goal."

^^ OK. Its my contention that reducing distortion in the reproduction chain leads to better fidelity/better sound, especially if the types of distortion we are reducing are the types to which the ear is particularly sensitive.

In this regard I favor sound quality over sound pressure (volume); I think you will find that most audiophiles do.

You can see in the specs of any amplifier (solid state, class D or tube) that the distortion is indeed higher driving a lower impedance load. If you think the distortion involved is negligible, its not, it usually is of the types that the ear cares more about.

This can be so significant that the use of an autoformer to raise the load impedance to an otherwise very capable solid state amplifier can result in improved sound. Were this not the case, the insertion of the autoformer would have adverse effects.

Here is the text of some communications that Paul Speltz (designer of the ZERO) has received from a well-known solid stage amplifier designer:

Hi Paul -

I hope you had a nice weekend out in the woods - it's cold out there!
I'm still enjoying the autoformers very much, and they are continuing
to improve as they break-in. I wanted to comment on why I finally
decided to try them (and wish I had a long time ago!). In the past, I
had always thought of autoformers as a "speaker tweak" or a kind of
crutch for amplifiers that couldn't handle difficult loads. I figured
that my amps could drive anything, so why worry about adding extra
boxes that I didn't need. Over the past several years, though, I have
been working more and more with line-level coupling transformers (part
of my new VRE-1 preamp design) and I have seen some evidence that
suggested to me that something in the nature of the load with the
transformers made the source driving circuits "happy," to use the
scientific term ;-) Thus I came to be interested in the autoformers as
a possible *amplifier tweak.* (I guess maybe you see them from the
speaker's point of view, while I tend to see things from the amp's
perspective.) Whatever the reason, they certainly work well, and I do
feel that the amp is "happier" with the conjugate load. Maybe it's
nothing more than a more benign impedance, but I suspect there is
something more at work. I suppose it might be possible to try them
with a high-efficiency,16 Ohm speaker as a test of some sort, but I
suspect that the sound would still improve with the autoformers.
Anyway, whatever the truth of the matter is, I am a convert.

Thanks again, Paul.

Happy holidays,

Steve McCormack
SMc Audio

03-26-14: Atmasphere
This can be so significant that the use of an autoformer to raise the load impedance to an otherwise very capable solid state amplifier can result in improved sound. Were this not the case, the insertion of the autoformer would have adverse effects.

OMG, it does, if the listener had ears. I've tried them on a very capable ss amp, and the sound took a bad turn for the worse.

Quotes: from the Dick Olsher review that was convienently left out of the excerpts of the reveiw of the link you provided on the Zero website, read between the lines, we all know that reviewers don't like to rock the boat too much.

"It is not a panacea, and as you can see from my experience, it does not always pan out. However, if you're the proud owner of an OTL or a low-power tube amplifier and presently driving a 4-ohm loudspeaker load, you owe it to yourself to give the ZERO a try."

Cheers George