LB: Let’s turn now to your preamplifiers, the Divinitive linestage and Phonatic phonostage. The first time I contacted you I asked about your phonostage, and you explained that you use opamps. I responded somewhat pessimistically. (Your suggestion for me to reserve judgment until I heard the unit was good one, as your phonostage is one of the best I’ve heard.) Why do opamps have such a bad reputation?
MK: Larry, thanks for the compliment on the Phonatic, but somewhat pessimistically? I don’t remember your exact words, but your tone made me feel like I should be out looking for some windmills to tilt. [Touche. LB]
When I began designing with opamps for hi-end, I had no idea of the stigma attached to them within the community. Opamps are the workhorses of the 21st century. Every electronic device from your cell phone to complex life saving medical equipment utilizes opamps. Just about every CD / DVD player, including hi-end units have at least one opamp inside.
Perhaps one reason that opamps are looked upon with disdain is the lies spread from the designers and manufacturers of SS gear using discreet circuitry (i.e., non-opamp design) that’s not unlike the lies spread by tube equipment manufacturers about SS gear. As the new kid on the block, opamps were initially subjected to ridicule, and possibly rightly so as early opamp designs were poor in comparison to their discreet older brothers. It was the same when early SS designs tried to compete with their older counterparts the vacuum tube. I think that as the youngest technology of the three, opamps are still growing and evolving. New design and manufacturing techniques are being developed almost daily and as the opamp matures and its architecture is refined you will see more great sounding designs employing opamps. The discreet SS has shaken its bad rap in all but the staunchest sects of tubedom and soon SS opamp design will become accepted as a valid medium for great audio design. The only thing holding it back now is ignorance.
For the first 30 years of hi-end audio, we had two factions, lemmings and lemming herders. The lemmings, also known as audiophiles, knew that they had something really good, but didn’t understand it fully and were overwhelmed. The herders were the periodicals and manufacturers telling the lemmings what was good and how they should spend their over-abundance of cash. By constantly introducing new products, each “better” than last years, and creating confusion regarding SS or tubes, analog or digital, isolation platforms, cables, power cords etc., the herders could get the lemmings to do (purchase) anything that they wanted. This was all done under the guise of education. But thankfully, over the years there were a few people on each side who would stand up and shout, “Hey everybody, the Emperor is naked!” And with each new insight, small “Were not gonna’ take it anymore” factions began to grow. The growth of these groups was greatly accelerated by the Internet. With free and instantaneous exchange of information around the world, people could get more opinions than just those of the herders. Today, we are all more informed, more educated and have more valid information on which to form a solid opinion. And with more people than ever shouting about the Emperor’s wardrobe, I am hoping that we are becoming more enlightened and more open minded, so as to accept and allow opamps to take their rightful place among hi-end audio designs.
LB: What are the main advantages (and disadvantages, if any) of opamps vs. traditional transistors?
MK: Opamps usually require less voltage and current than transistors and certainly far less than valves. They have a smaller footprint, so smaller circuit board and box enclosure. They use fewer raw materials in their manufacture, so they’re (technically) greener, and since the box can be smaller, there is less material used there also. There are many configurations from which to choose and customize your design.
They’re also temperamental and require careful implementation both electrically and mechanically. (PC board layout, grounding and supply decoupling.) And there are so many that it’s sometimes time consuming to choose the proper one for the design.
Regarding the quality of music reproduction, as far as I can see there are no advantages to using opamps, and despite what others are saying, there are no disadvantages either.
LB: Without giving away any trade secrets, what are some of the “tricks” for successfully implementing opamps?
MK: Tricks are for magicians, Larry. As with any design whether it be a bridge, automobile engine or audio reproduction device there are basic design rules and math that must always be observed, additional techniques such as close tolerances, component matching and layout that should be observed and quality control standards that will vary according to budgetary constraints and estimated retail price.
To take the basics a little further, when considering an opamp, we don’t just “Look” at those designed for audio applications. Taking a cue from Walt Jung, one of the great opamp gurus and pioneers of opamp design, we use current feedback (CFB) opamps in many locations. The suggested use for these devices is video drivers and transimpedance amps, but they do a great job passing audio when creatively implemented. If you want to build a better mousetrap, you might do well to build something that resembles a dishwasher instead of a typical mousetrap. Look at Viagra — it was originally designed to treat heart disease.
But if you really want specifics, my personal “Trick” has been, IS, and always will be a great power supply. Once you have designed a low-impedance, (as close to zero as possible) super voltage regulated power supply, the rest of the unit basically designs itself. This is especially true of opamp circuits.
LB: Due to the enormous levels of amplification required (especially for use with low output MC cartridges), every phonostage designer must confront the issue of keeping noise to a minimum. How did you approach this?
MK: Start with clean voltage from a really good…surprise; power supply. What I didn’t mention before is a supply that cleans up the line voltage by filtering out radio frequency noise and keeping common mode noise and EMI as low as possible. In addition, our printed circuit (PC) boards incorporate our own design grounding called a superstar ground.
We also select from the hundreds of possible opamps, starting with those with very low noise specs. As there are trade-offs with opamps regarding speed, bandwidth, current noise, voltage noise and a dozen other parameters that vary with each device, we perform SPICE (Simulation Program with Integrated Circuit Emphasis) simulations. These computer simulations help us quickly evaluate many different opamps and other components in design without building the actual physical circuit, saving time and money. Here is a good tutorial: http://www.ni.com/white-paper/5413/en
A real benefit of SPICE is that it can show us if something is bad; i.e. too much noise or distortion, errors due to math miscalculations, voltage levels and responses at various stages and much more. What it can’t do is tell us which, if any of the opamps (with acceptable parameters) will sound musical. It is for that extremely important reason that we also do extensive listening tests with many different prototypes to determine which of those opamps selected offer the best musicality in combination with each other in the various design stages.
In addition, when considering a certain design we are always looking at it from a “Less is best” perspective. The fewer components that pass the signal, the less noise and distortion will be realized. This is not to say that we would not include another stage of amplification as in the PhonaticMC and PhonaticXR (both of which have four stages of amplification when two stages could get the job done) simply because we don’t want to pass the signal through more components. Less means the least necessary to make a great sounding product. Noise and distortion specs mean very little if the design in not musical. I’ll take a slightly noisy LP over an MP-3 every time!
We use as little feedback as possible and in many design stages we employ current feedback (CFB) instead of voltage feedback (VFB). CFB opamps have faster slew rates and larger bandwidth in large signals with small to medium gains. They can also supply more output current making them well suited for output stages driving power amps. Both the Divinitive and Phonatic incorporate CFB output stages. For vinyl enthusiasts this means that the Phonatic can drive a power amp directly with low noise for cartridges of 1mV and greater with just a passive volume control in between. (Though I don’t recommend this set-up even with MA products.)
The Phonatic also incorporates several gain options that overlap in each successive stage. This allows the user to optimize the noise and gain for both the cartridge output and overall system gain and efficiency.
- ← Previous page
- (Page 4 of 7)
- Next page →
After reading that you would think everybody needs to hear these speakers & amplifier.