Why Vacuum Tubes?
By Douglas W. Fearn
(reprinted with permission from Pro Audio Review, Jan/Feb 1996)
For some of us in audio, tubes never really went away. We go back to the days when all equipment was tube equipment. (My first job, in the late 1960s, was at a Philadelphia radio station that was 100% “hollow state” from microphone to antenna.) I admit that I was thrilled when the first solid-state tape machines and recording consoles appeared, even though they didn’t sound as good as the old tube gear they replaced. The features were dazzling, so many of us were willing to tolerate a grungier sound, or pretend we didn’t hear it — at least for a while. (Sounds sort of like the digital vs. analog controversy now raging in some circles, doesn’t it?)
It seems that many of us who used tube gear in the past, and a growing segment of younger engineers and producers, have (re-)discovered tubes. Do they really sound better, or at least different? And if so, why?
Much (and maybe all) of the answer is contained in a paper in the May 1973 Journal of the Audio Engineering Society by Russell O. Hamm, Tubes Versus Transistors — Is There an Audible Difference? The quotes in this piece are from that article. Hamm found that the output level of studio microphones under typical recording conditions contains peaks far in excess of what VU meters display. Everyone knows that, but the peaks, as measured with an oscilloscope, are really high, easily exceeding 1 volt! The tube or transistor used in a condenser microphone, or in a microphone preamplifier, often will be driven into severe overload by these peaks. The peaks are short, so the sound isn’t grossly distorted-sounding; but the distorted peaks do affect what we hear. All preamps (and condenser mic electronics) are overloaded by these peaks, but tubes handle it differently than solid-state devices.
When transistors overload (in a discrete circuit or in an op amp), the dominant distortion product is the third harmonic. The third harmonic “produces a sound many musicians refer to as ‘blanketed.’ Instead of making the tone fuller, a strong third actually makes the tone softer.” On the other hand, with tubes (particularly triodes) the dominant distortion product is the second harmonic: “Musically the second is an octave above the fundamental and is almost inaudible, yet it adds body to the sound, making it fuller.”
There’s a lot more revealing information in Russ Hamm’s paper, and if you want more details about why tubes often sound better, it’s worth finding a copy.
Another difference between tubes and solid-state devices (including FETs) is the load they provide to the source. This is particularly significant with condenser microphone capsules. Even though an FET has an input impedance similar to tubes (in the megohms), for some reason condenser elements just sound better going into a tube. Is it input capacitance? Or some sort of dynamic loading factor?
If you ask enough people (particularly audiophiles), you’ll start to discover some truly metaphysical theories on why tubes sound better. Some explanations utilize quantum theory, some are simply wrong, while others are just plain bizarre. But in my own circuit designs I’ve encountered some odd phenomena that I can’t explain. For example, in a mic preamp prototype, I discovered that performance measurements were different if the unit was upside down. After I tried to eliminate all possible environmental factors, the difference persisted, even with a variety of tubes, and in different buildings. It was audible, too, though subtle. Was it caused by gravity slightly shifting the internal tube elements? The earth’s magnetic field? I still don’t know.
Tubes sound better because their distortion products are more musical. Tubes provide a more appropriate load to transducers. Those are the fundamental reasons why tubes sound better, but is there more?
