An Interview with Mark Kovach of Miracle Audio

V.             Active preamp

Price $500 and up.

Modern preamps have gains from 10 to15dB.  That means that your 2V DAC output can be as much as 10V at maximum output from the preamp.  When properly designed, this small amount of gain is not enough to add audible noise to the signal.  Input and output impedance stays constant for all components at all frequencies and listening levels. While it’s true that your 2V DAC output will be fine if driving a 400 watt amp into 92dB efficient speakers in a midsized room, just try the same DAC into a 60 watt tube amp into 89dB speakers.  In this scenario you’ll welcome the extra gain, appreciate the small amount of attenuation and hear the difference when the attenuator is “isolated.”

My first design (passive ladder attenuator) was the result of almost two years of extensive, controlled, and extremely well designed listening tests with the same ten people whose ears I trust implicitly.  There were Holco metal film, Roederstein mil spec metal film, Caddock  MK-132 and Vishay S-102 resistors in the tests.  The listeners could instantaneously select any of the four for comparison.  Based on the results, I built the attenuator, installed it and listened extensively.  The conclusion: Something was missing.

Sure the mids with female vocals were clean and natural, but the top was rolled off, slapped bass sounded sluggish and overall the music was not exciting.  I put a simple active unity buffer on the output and it was a new ballgame.  The music showed signs of life.  It was then that the passive attenuator was scrapped and the hard work began.  I experience this same lack of enthusiasm when listening to transformers and autoformers.  Even the very expensive models costing as much as and more than my top of the line DivinitiveXR have something that rubs my brain the wrong way.

The reason that I design speakers and preamps is that after years of listening and evaluations, I realized that they are the most important components of a system.

Speakers are the translators of electric into organic.  They turn the sub-atomic into the atomic, voltage waves into pressure waves and therefore are the most important aspect of reproduced sound and consequently the most difficult to design.

Preamps, and I mean active, are thought of as simply switchers and volume controls, but they are far more important than that.  They are translators and referees.  They’re analogous to the torque converter in cars, which efficiently couples the engine to the transmission in a smooth transition at all power and speeds.

Preamps are designed to efficiently couple front-end gear to the power amp.  Unity gain buffers keep the signal strong, preserving transients.  Active output circuitry supplies constant current to drive amps.  The convenient multi-input switching and volume control are side benefits.  In the case of the Phonatic and other phono stages, the three-fold function is to couple the cartridge to the linestage preamp, provide active gain and rewrite the inverse RIAA curve to correct for alterations during mastering.  In just about every configuration, the addition of an active preamp, with all of its additional circuitry will do more good than harm.

Some will say that their system sounds better with the DAC (or CD / DVD player) driving the amp directly; they may be correct for some of their music, or even all their music if their listening is limited to a certain genre or possibly even a limited area of a single genre.  I say that if you have a broad base and taste, then you are better off with a high quality linestage between all front end pieces (including a DAC) and your amp.

The only disadvantage of a preamp in the system is the expense of the added cables, and of course the added expense of the preamp itself, which in many cases is less than a step-up XFMR, autoformer or 48 step ladder with boutique resistors.

LB:      It is well known that volume controls can be a major source of distortion.  What type of volume control do you use, and why did you select it over others?

MK:     I’m not certain as to the amount of distortion that one can attribute to volume controls, but they most certainly can contribute to coloration of your audio.  Inductors as volume controls can wreak havoc on the dynamics, and stepped attenuators can exacerbate impedance mismatches causing lost dynamics and high frequency roll-off.

At Miracle Audio, the only component not designed by us is the volume control.  It uses an optical encoder to instruct a microprocessor to close relays connected to resistors which in various combinations create 60 steps @ 1dB.  It is designed by John Chapman and manufactured for us by his company ACL innovations.  We use this device because it complies with our high standards.  I believe that John sells this and similar versions of this electronic stepped attenuator to about a dozen other hi-end manufacturers, so I’m guessing that we all can’t be wrong.  For more info go to: http://www.bentaudio.com/index2.html

LB:      What else can you share about the overall design?

MK:     As mentioned previously, the preamp should be a referee between power amp and front-end components.  The Divinitive’s active circuitry combines VFB and CFB opamps to deliver a stable, quiet output with plenty of current to drive just about any power amp.  This takes care of what’s ahead (power amp) from an impedance standpoint, but that’s only half of the solution.

Because the ACL attenuator (remember I mentioned that active preamps have passive attenuators) is before the active circuitry, the front-end component can’t look directly at its (the attenuator’s) input or you have the same problem as using just a passive attenuator; it’s still half as bad because of the benefit of the active circuitry.  To isolate the attenuator in the Divinitive from the front-end component, we employ a unity gain buffer, again using VFB and CFB opamps.  It’s a scaled down version of our output stage with some secret sauce that differentiates it from most buffers in use.  This takes care of the front, and now we have complete isolation of front-end component from power amp insuring against any possible impedance miss-matches, and insuring that your signal will include all transients without tonal colorations.  I feel that most systems utilizing only passive attenuators and transformers could benefit from the addition of a simple input buffer, and benefit greatly with the one in the Divinitive line stage.

For convenience, we’ve included front panel controls that duplicate most functions of the remote (in case your dog eats it), and we also offer several options that can be ordered initially or added as factory mods at some future time.  These are:

1. True differential balanced output driver:  This line driver incorporates four channels of identical active circuitry assuring highest quality output for the inverted and non-inverted signal.  You can also use either single ended RCA or Balanced XLR by simply plugging into the appropriate jack; no switching needed, so fewer contacts in the signal path.

2. A high-pass analog filter:  If you use a sub with your system, this analog filter allows matching the cut-off frequency of your main speakers to that of your sub, without routing the main signal through the sub’s circuitry. You can also use it to attenuate unwanted low and sub-sonic frequencies that your speakers just burn off as heat, making them more efficient, even when not using a sub.  It includes a 2 pole and 4 pole slope and adjustable cut-off frequencies.  It also employs VFB and CFB circuitry, all in the analog domain.

3. Sub /Aux line driver: Instead of splitting the RCA output to drive a sub, or utilizing the extra pair of parallel jacks, as is standard on many preamps, an optional full bandwidth stereo line driver can be added.  This card has its own buffer and tracks the attenuator identically to the main outputs.  It can also be used to drive a whole house audio amp.

Larry, just briefly, I’d like to stress the harm, sonically that is, that one does to the main signal (and sub signal, for that matter) when you split or parallel your main outputs to drive a sub amp and main amp together.  Although opinions vary as to sonic quality, it’s a fact that the preamp’s output voltage splits causing a drop of 3dB in the main signal.  This requires more gain to bring it up to the previous level and that means more noise — period.  The current also splits, and as far as I’m concerned, if you want dynamics, you can never have too much current when driving power amps.  And last but not necessarily least, the impedance is calculated as ((amp A) X  (amp B)) / ((amp A) + (amp B))  This means that the total impedance that your preamp “sees” is now less than the lower of the two amps impedances — in some cases too low to maintain the minimum 10 to 1 ratio that in my opinion should be 100 to 1.  This scenario could be worse than using a non-buffered passive attenuator from an impedance standpoint.  And audiophiles wonder why their system sounds awful when they add a sub?  Surprise!  It’s not necessarily the sub.