Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Observations and expectations

One of the pitfalls of being an evaluator is presupposing the results. For example, if you test drive a new car and think “I am going to like this car” then you may miss some of the problems you might otherwise have found. This happens because you cloud your judgment with expectations of what you think will happen, rather than being the careful observer. And worse, when we precondition ourselves to think one way we then believe the conclusion if it should suit our purpose. For example, this morning I was afraid to get on the scale and weigh myself for fear I would not like the results. But I forced it and was pleasantly surprised. My immediate reaction was to believe the result I had hoped for and never questioned if it was but an anomaly (something experience has shown me it probably is).

Easy observation to make, difficult to eliminate prejudice in practice. This same issue comes up when evaluating high end audio. We all have to be careful when bringing in a new piece of kit, evaluating new firmware, cables, music, or room treatments.

If we think it might be better and it is, case closed! If it isn’t what we expected, we then dig deeper, something we should always do anyway. It’s a good reminder to each of us to take a deep breath and let the initial euphoria of expectations matching observations and make sure they are accurate.

Certainly for me.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

It’s ok to make the wrong call

Critics are people whose opinions are valued by others, and we’re all critics to someone. Perhaps your children look to you for approval on food, movies or entertainment. Or maybe it’s your contemporaries with whom you look for guidance. Or maybe you read and contribute on a blog of like-minded people who offer their evaluations of music or stereo that help guide your decisions.

The greater number of choices we encounter, the more we rely on others to help us select from the list of options. We simply cannot know all the music available, all the stereos to purchase, all the restaurants to eat at, all the political party to back, all the computers to purchase, or all the places to visit on vacation. There are just too many choices. Things weren’t quite so complicated even a few years ago and yet… As the roles of critics becomes increasingly important, many find greater comfort and safety in following their suggestions rather than making their own choice. But there’s a downside to this route and most critics worth their salt will tell you so. The more you abdicate your role of decision maker, the greater the chance of isolating yourself from failure. And failure is as important as success, perhaps more so.

Be careful of working hard at not failing. It is failure that shapes us more than our successes.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Sommelier of the ear

I watched with fascination as aspiring sommelier’s identified wine types by tasting. Not only did they accurately identify year, country and vineyard, but in some cases much more. And all from tasting the wine in a blind test. Sound familiar? Most Audiophiles I know of have a similar set of skills that rely on all the senses and a great deal of experience. In both cases these experienced tasters and listeners use their internal measurement systems to determine that which they are experiencing.

There are many that still believe the ability to identify characteristics in sound that we cannot yet measure is a hoax.

They would refer to those that acknowledge this ability as ‘Audiofools’, a derogatory term from ‘Audiophile’. By coming up with such a disrespectful label, others on the periphery of knowledge assume it ‘says it all’ and move on to the next controversy. That is a shame because there is a chance for all of us to learn from those experienced enough to both identify a wine simply by taste, or discover musical truth by listening.
Until several years ago there were no machines capable of doing what an expert wine taster can do. Does that negate what they achieve? No, because the evidence cannot be refuted.
We still do not have the means to measure that which we hear, yet there are many who would steadfastly deny that we hear any differences at all.

In the first case it is easy to understand why people who might normally deny that wine experts can do this relent: the evidence appears on a bottle of wine for all to see. The second case is equally difficult to believe in, yet no less real.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Amplifier artforms

Sometimes that which I write can conflict with my world view of what we do. This happens when I am trying to make a point about one thing and unintentionally is at the expense of something else. A good example was yesterday’s post where I suggested a vacuum tube was the last thing I would design into an output stage of a power amplifier, yet the first thing I would add to its input. It was assumed by some readers this means I do not like tubes nor the work of tube manufacturers building amplifiers. And that is not what I wished to come across.

Some of the finest sounding power amplifiers I have ever heard were completely tube based. They are hard to beat with solid state means. Which is why we have decided to add a tube to the front of our new amplifier set for soft launch in April. But the fact that I think using a tube as an output device in a power amplifier’s a bad idea doesn’t mean I don’t enjoy the results when others do. No, to the contrary, I do enjoy them and tolerate their shortcomings. And every design has shortcomings.
Take damping factor as an example. DF is a measurement of how well a power amplifier is able to control a loudspeaker. The higher the DF, the greater the control over the speaker. Tube amplifiers have less DF than their solid state brethren and exhibit less control over loudspeakers. How much this affects the sound quality of what we listen to has much to do with the type of loudspeaker and one’s sensitivity to the results. A highly inductive dynamic loudspeaker requires a lot of control in order to have authoritative bass response, a tube amplifier struggles to provide this. A solid state amplifier with adequate design hasn’t any problem in the same situation. And this is just one example.

Designing high end audio equipment destined to tickle the pleasure centers with well reproduced music is an artform. There are no perfect solutions and many paths to achieving great sound.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Food for thought

One of my readers asked me an interesting question recently concerning the appropriate time to make the transition from analog to digital. Should you do it as soon as the digital audio comes from the source, as you would in a DAC? Or would it be better to keep it in its native form as close to the loudspeaker as possible; decoding it into analog in the amplifier instead?
There are good arguments for both. Converting to analog in the DAC means that everything else in the chain can also be analog, like phono and tuner, as well the preamplifier. This is the right choice if you have a mix of analog and digital sources in your chain. If, instead, you have a pure digital chain you might be better served keeping it in its original format until the very last moment.

Few power amplifiers today have digital inputs. This is for several reasons, but chief among them is limited commercial appeal. Imagine purchasing a power amplifier with only a digital input. How would you playback vinyl? There are solutions, such as the NPC, a hybrid analog/digital phono stage you can use for this very function, or to build an all digital system. Yet I think most Audiophiles are still more comfortable with analog centered systems.

It seems premature to consider an all digital system as the norm, yet…as we continue down the rabbit hole of our changing technology, the shift towards analog as a secondary medium continues to inch forward. I wouldn’t be surprised to see the day when most amps have digital inputs instead of what we have now, which is the opposite.

Food for thought.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Make the right choice

Thank goodness we have choices in the devices we design into an amplifier; not all are best for a given application.
If you were to ask me to list the worst devices to build a power amplifier output stage a vacuum tube would be first. And there are a number of reasons why.

Let’s remind ourselves what the task at hand is: moving a loudspeaker driver back and forth in response to a musical signal. A small change at the input is translated to a larger change at the output. Or said another way, a power amplifier uses a little bit of power at its input to produce a larger power at its output. This occurs when the small input power supplied from the source of music connects the loudspeaker to the amplifier’s power supply in varying degrees. It is the same concept as controlling a water faucet: a small change to the water valve results in a big change at the output of the faucet. And it is the action of the aforementioned valve I am concerned with.

Connecting the power supply of an amplifier to a loudspeaker demands an accurate, proper valve. The best device for the task is one that provides a high input and low output impedance, preferably operating in the current, rather than voltage domain. Tubes are best as voltage amplifiers, transistors as current amplifiers. It’s why in the upcoming BHK Signature we opted to choose the device best suited for each activity: a tube in the front end for voltage amplification, transistors in the back end for current.

In my research for similar design concepts I found a few attempts at similar hybrids. Surprisingly, few got it right. Most flipped the formula on its head in a bewildering fashion: transistors on the input, tubes on the output. Bizarre.

I am convinced some designs exist to be different while others strive for perfecting musical truth.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Class B, AB, and A

Ever wonder what the classes of amplifiers mean? There’s Class A, Class AB, Class B (and a few others). Here’s a bit of history and an explanation.

Early transistor power amplifiers had their share of problems: crossover distortion, poorly designed output stages, slew rate limitations, transient slewing, little attention paid to power supplies. These failings of solid state amps fueled the tube devotee’s negative feelings toward them; and rightfully so. Until these issues could be fixed, few if any tube people would consider moving from the problems of tube amplifiers because there really weren’t any better solutions being offered. Things would change, but not for a little while.

Perhaps the biggest failing of early solid state designs, aside from their tendency to occasionally self ignite, was crossover notch distortion; a problem easily overcome by increasing the level of class A bias. Crossover notch distortion was fairly common amongst early solid state amplifiers that took advantage of the fact transistors had two types: positive and negative. Tube designers had only one type available to them: positive going tubes. The fact that solid state amplifier designers could use a positive transistor for the top half of the waveform and a negative transistor for the bottom half removed the need for a special inverting circuit some tube amps and push pull solid state amps used, plus benefited from an increase in efficiency. Less circuitry meant better sound. Only, if the two output devices were not used properly, this new type of bright and harsh sounding distortion would pollute our music.
Here is a picture of severe crossover notch distortion.

This happens when the positive transistor shuts off before the negative transistor turns on. This type of output would be called a Class B output and was almost never used in any quality audio products. However, the next scope photo was more indicative of what actual crossover notch distortion looked like.

Note the breaks in the middle of the waveform. This area is heard as a harsh and bright coloration to the music, can be rather jarring and was fairly prevalent in early power amplifiers. To fix this problem we would turn each of the two transistors on just a little bit. This is called Class A biasing and the amount you turn it always on determines how we classify the amplifier: no bias and it is a Class B amplifier. A few watts of constant bias and we refer to it as Standard Class AB bias. Both transistor always on, even if there is no signal, full Class A. Early amplifiers, as well as many today, run at a few watts of class A bias; just enough to eliminate the crossover notch and no more than that.

The net result is three types of amplifiers: Standard or High Class AB, Full Class A. To tell, it’s reasonably easy. Just put your hand on the amplifier’s heat sink after it’s been idling with no music playing. If it’s just warm, probably standard Class AB bias. Good and hot, yet still able to keep your hand on the sink, High Bias Class AB. Hotter’n a firecracker, Class A. Here’s a picture of a true off-the-chart fully class A amplifier from our friend Nelson Pass.

So now, when you see a solid state amplifier manufacturer say they have a ‘Class AB amplifier’ (as most do) you can ask the obvious question. “How much class A bias?”

In the case of our new BHK Signature power amplifier, slated for release this spring, the answer is 40 watts per channel (High, Class AB bias). Which means that for the first 40 watts of power delivered to the loudspeaker, the amp is a true class A device. Since most loudspeakers play quite loudly at 40 watts and the majority of your listening is under 40 watts, the amplifier is essentially class A for all quiet to medium loudness music. Peaks and large dynamics are then not Class A, but since most of the benefits of class A bias are heard in inner detail and quiet passages, there’s no need. Simple, eh?

Now you know.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Turning the tide

Once the transistor had been released to the general engineering community it didn’t take long before those that used it stopped trying to duplicate tube circuits and ventured out into the new design idea phase. At that time tube designers were still the majority and in audio circles they reigned supreme, unaware of the impending avalanche of solid state amplifiers that would soon overwhelm their business.

One of the first innovations transistor designers discovered was how to eliminate the audio transformer prevalent in tube designs. It was easy enough to replace transformer with capacitors and then a few brave souls realized they could actually replace the capacitors as well. This technique is called Direct Coupling and is still in use today in better sounding audio designs. The elimination of transformers and capacitors contributed mightily to the improvements in sound that progressed along this circuitous path. Direct coupling an amplification circuit from input to output is nearly impossible with tubes, but transistors did it easily. Things progressed on a reasonably predictable path for a few years, with American and European audio companies tinkering and toying with transistors. The Japanese, however, were embracing solid state electronics with a vengeance and that work would soon wash ashore in the form of a market onslaught that would forever change our industry. But something else was brewing that was even bigger.
It became obvious to engineers and physicists in the nacant semiconductor industry that the real power of the transistor lay not in a single replacement for vacuum tubes, but miniaturizing and integrating many of them into a tiny circuit. It was discovered that the very materials used to make transistors could also be used to make the associated components that transistors relied upon to perform their various tasks: resistors, diodes, capacitors. Soon these pieces of the great puzzle came together in what we now call an integrated circuit; a small postage stamp sized piece of silicone that could take the place of an entire PC board full of parts. The IC was born and within a few decades everything changed.

Of importance to our story of how it changed amplifiers within our industry, we turn to a truly crazy bohemian of a man; Bob Widlar. Widlar, a graduate of our own Boulder Colorado university, worked for a time at Ball Aerospace company, just around the corner from PS Audio. Widlar managed to have what I consider the single greatest impact on modern audio design since Lee De Forest. He is credited with hundreds of inventions but among them and important to our industry: the first mass produced audio op amp, current source, integrated voltage regulator, audio output stage, FETs to op amps, and the list goes on. PS Audio’s very first product, our phono preamplifier, used a Widlar designed IC known as the 709 op amp, the only op amp of the day that sounded good. No, sounded great!

Once the Japanese manufacturers got hold of ICs all hell broke loose and the race was on. In high end audio tubes were still king, but a budding number of companies using solid state began to emerge, challenging the status quo. PS Audio was amongst those pioneers.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Early amplifiers

I mentioned in yesterday’s post that the early transistor amps invading the hi fi and consumer electronics scene were dreadful sounding. But why?

The tale is long and complicated, filled with many twists and turns and the complete story is not within the scope or space of these short daily posts. But I can shed some light onto the subject, light that will help us understand what happened and some of the evolutionary process manufacturers went through from both sides of the aisle. And by the ‘aisle’ I do mean the same definition as the common usage of strong political sides being taken. In Washington those sides of the aisle can get pretty polarized, sometimes downright nasty. This is because the differences of opinion of those in congress can be very much opposite of each other. And such is the case with the two sides of the aisle in audio reproduction: those that listen as the final test and those that measure. But I jump ahead of my story.
In the beginning of transistor amplifiers it was common practice to simply replace the vacuum tube with a transistor and leave everything else the same. What you wound up with didn’t have the frailties and requirements of a vacuum tube (heater power, microphonics, warm up, relatively short lifespan, heat), but also lacked the sound of the tube; a sound that had been honed over many years of work and listening. The reason this was done was simple, the engineers who implemented the new fangled transistors were schooled in tubes; they knew no better. It’s instructive to remind ourselves that the pioneers of solid state electronics at Bell Labs were trying hard to replace the vacuum tube as well as the mechanical relay that made up the massive phone system in our nation (and those around the world). Imagine how expensive, how difficult it must have been to handle all phone switching and amplification with the likes of vacuum tubes and mechanical relays. It took armies of technicians to replace tubes and relays that routinely failed and these antiquated devices limited the performance of the system as well. To send telephone calls over long distances you needed many repeater stations along the way, each powered by tubes and relays, devices ill suited to the task.

The pressure was on to replace the glowing fire bottles that plagued Ma Bell, and a team of physicists headed by one hell of a crazy man, William Shockley, led the way to developing the transistor that would forever change not only the phone system but everything else in our lives; and not by just a little bit. So it was not by accident that the engineers of the day, including those of audio companies, did little more than replace vacuum tubes with transistors and call it new. But with any new technology, you need to branch out and take advantage of what that device has to offer, something these early engineers did not do. I can remember gasping at these circuits in horror. Here’s a picture of one such approach.

The little curly symbol separating transistor Q1 from Q2 and Q3 shows a transformer between them. And the output of Q2 and Q3 also has an audio transformer to couple to the next stage. This circuit is guaranteed to sound awful. I do not need to prototype it to know. It’s just bad design. But that’s what we had in the beginning. Tomorrow we’ll move forward as a new crop of engineers learned to take advantage of the transistor’s unique characteristics that make it superior to a vacuum tube in some applications.

As an aside, reader David Blank sent me an interesting tidbit of info worth sharing. Turns out one of the first Pacific Stereo stores (if not the first) was in Palo Alto, in the same building as Shockley Labs, one of the three men credited with inventing the transistor.

Asheville, North Carolina ‘s Home Theater and Audio specialists present Cane Creek AV and Paul McGowan – PS Audio, Intl.

Audio tsunami

My friend Jack Bybee once told me the only result of being first in the marketplace is to be the recipient of many arrows in your rear end. In his opinion, bringing new technology to market rarely benefits the brave soul that introduces it. Instead, it’s better to wait until all the problems are worked out and then launch your own version of the new technology that benefits from not making the same mistakes as the early versions. Jack’s a wise man, although he preaches a lesson I rarely paid attention to. To this day I still bear the scars of a few arrows on my backside.

The truth is, rarely do first attempts at new technology get things right. Transistor amplifiers are a great example of this observation; the first lot of them were beyond dreadful sounding.

Imagine a world of vacuum tube amplifiers; their designs honed, their sound perfected over a fifty year span. Into this world of great sounding audio, throw a poorly designed transistor based power amplifier, proclaim it as the next ‘thing’, and then run like hell. The fireworks, criticisms and arrows aimed at the unsuspecting company, would fly with merciless force and sting with every point of contact. It’d be ugly. It was, in fact, ugly. But it did not stop the march of transistors from coming.
Into this hotbed of the audio market they came; first in drips and drabs, then with the fury of a tsunami as the Japanese electronics manufacturers entered the worldwide marketplace with a product they popularized, know as the receiver. Their biggest marketplace would turn out to be the Americans, but not those of us in high end audio through our hometown shops. Instead they found sanctuary and success overseas with the hundreds of thousands of US military. The Vietnam war and the troop buildups remaining in Europe during the late 1960′s and early 1970′s were hotbeds for young music lovers with money in their pockets. An entire industry flourished overseas, its influence spreading across the globe and into the US, engulfing the high end audio market. During this era people like Richard Schram, now the head of Parasound, rode this wave of Japanese and American receiver takeovers in the US by opening up some of the first chains of audio stores. In Richard’s case the chain was called Pacific Stereo.

So what made these early transistor amplifiers sound so bad? And what changed to turn the tide and start pounding nails in the coffins of the thriving vacuum tube amplifier market? Over the next few days we’ll start looking at some of the causes and results.

Stay tuned.