I am kind of semi-evaluating a power amplifier now and think about the following: there is a correct relation between harmonic content of amplifier and it’s dynamic aptitude.

I have heard many amplifiers that run across the dynamic range with a tremendous speed but the while they do so are loosing the “harmonic tails”. As the result, while they operate in transient they sharpen sound (and the Morons-audiophile love it, it is what the call resolution!) Probably the most flamboyant amp that does it would be Canadian Tenor OTL that keeps thrilling the audiofool intelligence. They, and the similar amps, combined with dull and “slow” speakers and playing essentially simple audiophile music mask out the screwee dissociation between own dynamic qualities and speed of it’s exchange, at least it hapends in the eye of an typical non-demanding audio beholder….

I can see some hard-core technical people coming and would suggest to analyze the distortion patters and to correlate between the input and output signals at the different volume levels. I do not find this approach being indicative and I have 2 reasons ( I learned it designing the loudspeakers). First, when I do so I get a static presentation of harmonic capacity. I do not really care about it and I more concern about the amp’s ability to maintain the acoustic-like harmonic stricture while it is in a move. Second, and this is sound kind of ridicules: I always, always-always, prefer the “fatter” harmonic, the larger the better.

Interesting that I came across to an observation that the acoustic-like harmonic constraint is NOT my reference while I try to assess the performance of audio. Rather I am searching for a “believable”, reasonable and justifiable balance between the MAXIMUM harmonic load and the speed of its exchange. Therefore within my own language I always recognize an expression  - “HARMONIC SPEED” which represents in my mind a subjective relation between a speed of sonic messages and this maximum harmonic content. The important thing in here the word “maximum” as I NEVER heard any EXCESSIVE harmonic content from audio but rather I heard the insufficient speed of manipulation of harmonic content…(add to here the dynamic compression and you get how audio generally sound)

Anyhow, I do not pitch any solution but just thinking out loud about the subject. I really do not know what in an amplifier responsible for speed, acceleration, velocity and so on. I kind of know what it might be but I have no idea how to change it and in the same time to assure that the amp will maintain its MAXIMUM harmonic puffiness. I know how to hear it and to assess it but I do not know how to manage it, at least deliberately….

As the result, I feel that we get the “harmonic speed” in our amps “as is”. One amp has it better another worst. If it were up to me I would introduce some kind of standard in there that force the manufactures to comply with the speed/harmonic-load ratio. If the amp has too fast speed for a given harmonic stricture or too striped down harmonics for a given speed then I would classified the amp as “audiophile products” and would demand it to be sold with a label:

 ”Attention: the given product has an uncivilized harmonic speed. The sever long-trim listening consequences might be inflicted by using this amplifier,”

Rgs,
The Cat

Duke, you see, the lowering of output impedance is great but once again “how it was used”. When we hear the result do we hear the effect of a more infinite dumping or we hear the problem that coming with application of feedbacks, or we hear the problem with the fundamental fault of the drivers?

I few years ago a friend of mine (Geoff Cook) turned me to think and to experiment with negative output impedance when a negative output impedance cancels out the mechanical and electrical suspension of the LF drivers. I would not comment if the lower Zout sounded better or worth.  There are many conditions to consider starting form the way that the amps used and ending with the specific characteristics of the drivers/enclosures. I found at that time (I think it was 2001) a guy – he was professor in some kind of university in Mid West who wrote in beginning off 90s his PhD papers about a contribution of an amp’s negative output impedance for linearizing of the LF drivers performance. I spoke with him a few times and it was hugely educational. All this made me to feel that a dropping out impedance is a hugely powerful tool if it properly used. You might see a few commercial products from 80s that use it but it was not properly done and it was in context of the very wrong drivers.

You see, the biggest and the most magnificent thing when you introduce an infinite damping with lower or negative impedance is getting rid of the damn primary resonance on the driver. We all sucked in the wisdom of building LF sections heavily relaying on the damn Fs. But the primary resonance is a fundamentally crappy thing! The impedance do up, the amplifier’s output stage become instantaneously underloaded and the cone moves not according to the signal but according of God know what (mechanical suspension of the driver, air damping and so on…)

Audio people are not familiar with bass reproduction when the Fs do not exists and I feel that it is very unfortunate. My experiments did indicate that with no Fs we get completely different type of bass, the real bass! I have to admit that I did not get better sound in context of the speakers and the amplifiers that I used. You see, I did not listen the Sound itself but rather the specific strictures of the sound and that lead me to my conclusions. In fact in a few weeks I am planning to return to those experiments with the LF channel of my “Super Melquíades” project and to see if I would be able to get an “interesting” result by dropping the output impedance to approximately minus 2Ohm. I consulted with Dima and we sumized that with Melquíades I would not be able to cancel the driver impedance completely because the Melquíades has no enough power (my former experiments suggested to use minus 80% of DCR). However I would be able to get a “minus little bit” and will see where it will lead me.

I think it should work fine. When I experimented with the negative Zout in past I used different (wrong) woofers. Today, although my woofers are not underhanged but they are very “smartly” overhanged and the coil in it’s Xmax still does not jumps across the magnetic transient. I pay a fee of having no sensitively but I have many drivers… and ... a “beneficialy” small room. My preliminary experiments indicated the Melquíades could loose approximately 4-6dB and still drive my room to the necessary loudness. In addition I change the gain in my line-stage from 3dB to ~8dB (thanks, Dima). So, I will have some Melquíades power to burn in the dumping…

When I reach some conclusive results I will post them around here….

Duke, here are are some observation for you to think.

Duke wrote:

The soundstage collapses; the presentation becomes dry, lifeless and even edgy; and any emotion in the music is completely lost.   When the output impedance control is dialed back up (damping factor down), the soundstage blooms, the music becomes "wet", and the emotion in the music begins to move you again.

What I’m hearing in your description is your drivers are flatting out the primary resonance. If you do not use any barbarian impedance normalization network (that must not be used anyhow) then you might easily detect that with dropping output impedance the rise of the speakers impedance become less aggressive. (With the conventual woofers) So, what the Morons (means we) recognize as “bass” is in fact a situation when an amp begin to pump hundreds presents of its typical power to the drivers that temporarily lost a control over themselves. With the Negative Zout since the driver’s shaking begun to be slightly reduces via the force of the amp’s output stage then you loose subjectively what we call “bass”. Keep dropping the impedance while you run your real time spectral analyzer and you will see the actual result of bass disappearing (actually it is not bass disappearing but the amplitude around the Fs)…

Anyhow, any single adjective that you used in your description of the effect you heard is an exact representation of a gradual removing LF component from Sound. So, if you want to hear the clear effect of damping increasing with drooping of Zout then you should in parallel with dropping of Zout to EQ the LF up.

This of course introduce a whole bag of own problems…

Paul S wrote:

Ironically, the "best" musical bass I have gotten is from the ML2s 16 Ohm taps, which seems to fly in the face of both the concept and the obvious problems that the Z/load creates.  What I am saying is that despite the "kitty's furry belly" problem, the harmonics and general musical character of the bass is either a separate issue or there are other forces at work in this equation that I do not yet understand. Given the truth that inheres in your observations, Romy, it still seems like Zarathustra would be the most fruitful path.

I do not know if it would be accurate. I am reviewing now my view of bass reproduction.  Something is very interesting in single-ended bass. Usually people look at bass they see quality of max amplitude but what most interested me lately is to look at bass as low amplitude. An amplifier is like a 16-bit D/A converter – at minus 10dB you have good signal but as long you dive at minus 80dB you begin to see garbage. The PP amp might do impressive loud bass but SET what it very interesting is the bass that a SET doe at 5% of own power – very special feeling….

Still, I do not how to look at it farther - I never finished the Zarathustra II and I do not have any other “better” amp. At this point I do not look for anything better of different as the LF channel of Milq it looks like an handle my room. I am non-violently looking at a big ass-amorphous core bass transformer but I do not know if I go there as it would require putting second order filter on bass…