I have always noted with great curiosity and interest the fundamental and the rather consequential significance of sonic characteristics between transistors and vacuum tube circuits.
My personal inclination is I prefer the gestalt presentation of tubes, since the impression they convey in comparison to solid-state is a better sense of harmonic structure, palpability and bloom, as well as a more natural impression of scale and timbral and harmonic accuracy. My engineering sense didn’t agree with this assertion, saying that, “objectively” speaking, there is no particular reason as to why I would prefer vacuum tubes over solid state. The only engineering parameter I could conjure up in support of a preference was that tubes generate their non-linear behavior as second harmonic distortion. But is the human ear very sensitive to harmonic distortion on the order of only, let’s say, .01%, which is achievable in both sold state and tubes?
So, I decided to do some investigating on my own into this old age. Let’s face it, the “tube versus solid-state” face-off is still one of the most prominent, defining paradigms in high-end audio, and that amounts to science being unable to correlate and substantiate what can be easily perceived. The exactitude of science is being questioned and this has been one the greatest controversies of humanity for as long as science has existed. To me, the dichotomy is an invitation to inquire deeper, and to approach problems with new approaches and techniques. “The answer is out there”, all we have to do is find it.
My own research indicated that the most correlating and objective parameter between what we hear and measure is the measurement technique called transient intermodul-distortion. This technique evaluates an amplifier system under dynamic conditions, thus more likely to correlate what we hear with measured response. I have expanded upon the normal convention by investigating the internal square wave response in each point inside the circuit. To illustrate, the success of this test method has been useful. My partner, Rick Schultz, implemented an invention of his: “Speed Of Light”, or SOL for short.
This innovation uses magnets placed adjacent to an audio circuit to enhance the flow of electrons through all circuit components. He implemented this enhancement on a KST-150 power amplifier. The difference I perceived in the sonic quality was startling. Much improved. Curious, I then tested the circuit with SOL using the aforementioned square wave test and compared the results with a stock KST-150. The results showed the amplifier incorporating SOL had a noticeably better internal square wave performance at the input stage. Referring back to the primary consideration of using tubes, when I perform a square wave test on a tube circuit, I get a cleaner waveform. Why is this so? Well, it just so happens that “inside” a transistor, there exists a POSITIVE feedback loop called “hre”. This is incorporated intentionally to increase the gain or the amount of amplification a transistor provides.
You see, back in the 1950’s when transistors were in their infancy, they had very low gain and were generally unusable for this reason. It wasn’t until the mid 1960’s when useable gain transistors came onto the market, and transistor manufacturers used hre(positive feedback)to obtain this enhancement. Of course, when they did this it compromised device linearity considerably, and since linearity is the most important and relevant characteristic pertaining to audio circuits, so when the first solid-state amplifiers appeared on the market place, perceptive listeners condemned the sound quality using adjectives like grain, harsh, gritty, edgy, etc. Today, transistor manufacturers have designed device topologies which don’t use hre so heavily to get sufficient gain from solid-state devices; but, it’s still there to some degree. Hence, there continues to exist identifiable sonic characteristics when transistor amplifiers are compared with those of tube.
So, from a sonic and device linearity standpoint, tubes offer superior performance which can be quantifiably verified using the square wave test. So if I were to go with sonic performance as a primary guideline, I would choose to go with tubes. From a manufacturer’s standpoint, there are logistical problems with tubes, like longevity, proneness to damage from physical shock, generation of heat, and vibration-introduced microphonics. Those are the primary problems. Since this is a pre-amplifier circuit, these considerations are minimized, as the signal amplitude is “low level”. Still, the problems are there.
Last but not least are the commercial aspects of tubes, like market appeal and cost. All of these considerations must be accounted for when making a determination of which to go with: tubes or solid-state.
In my next installment, I shall go into a detailed circuit design description of the tube circuit, and the important design parameters I utilized with the solid state circuit.
- (Page 1 of 1)