My vinyl is pristine also, and I did the same thing - recorded on a Nakamichi (which I still have) with Dbx encoding first thing and then played the tape from then on except for special occasions. I still have them, but I listen to CD sound now - the noise from even immaculately cared for and cleaned vinyl was enough to compromise the immersive experience of listening for me, and I really appreciate the greater DR and channel separation, but that’s a personal preference just like tubes vs solid state amps/preamps. It’s all good - whatever you enjoy and makes you happy. As I wrote in the introduction to the Euphoria loudspeaker: “if your current system doesn’t move you to tears when you listen to the music you love, you’re listening to the wrong transducers”

CD's have been accused of sounding "sterile" when compared to the "warmth" of vinyl. But like you, I like CD sound a lot more than vinyl or tape, except for one thing: Album covers!

I started transferring vinyl to AIFF files for audio CDs years ago, but found it tedious to separate the tracks so the CD could be indexed and accessed properly. And like you, the surface noise was annoying.

As for .mp3 and AAC files, what really makes them sound bad is a low SAMPLING RATE. It's like audio streaming on the Internet. The lower the sampling rate, the smaller the file and the less bandwidth required, but below about 256 Kbps, the quality gets annoyingly bad. So I only listen to NPR and BBC News on the Internet at home. When I copy my own source audio to AAC/.mp3 formats, I choose 384 Kbps. When I buy digital tunes, I get the 256 Kbps version.

That said, 44.1 KHz AIFF files on CDs and 48 KHz video sound tracks are really fine. I stick to them when I can.

As for "that tube sound," it is real. Tubes create harmonic distortion that is both musical and desirable. When the signal gets loud, the distortion builds gradually, but the signal doesn't clip harshly as it does with transistors. That is why musicians still use tube amps for guitars and bass. They can control distortion at the preamp. The best single input tube pre-amps for studio and stage can cost $1800 new. Good guitar amps are hundreds of dollars, and better ones cost thousands. Musicians treasure amps from the 1960s for their vintage sounds. They spend small fortunes getting them refurbished with new capacitors, controls, safety upgrades, and speaker cones.

In audiophile tube systems, the distortion is very low, but what is there can be important to the overall listening experience. It adds some of what the French call, "Je ne sais quoi". The idiom means "I don't know what (that is, but it adds something I like)."

CD's have been accused of sounding "sterile" when compared to the "warmth" of vinyl. But like you, I like CD sound a lot more than vinyl or tape, except for one thing: Album covers!

I started transferring vinyl to AIFF files for audio CDs years ago, but found it tedious to separate the tracks so the CD could be indexed and accessed properly. And like you, the surface noise was annoying.

As for .mp3 and AAC files, what really makes them sound bad is a low SAMPLING RATE. It's like audio streaming on the Internet. The lower the sampling rate, the smaller the file and the less bandwidth required, but below about 256 Kbps, the quality gets annoyingly bad. So I only listen to NPR and BBC News on the Internet at home. When I copy my own source audio to AAC/.mp3 formats, I choose 384 Kbps. When I buy digital tunes, I get the 256 Kbps version.

That said, 44.1 KHz AIFF files on CDs and 48 KHz video sound tracks are really fine. I stick to them when I can.

As for "that tube sound," it is real. Tubes create harmonic distortion that is both musical and desirable. When the signal gets loud, the distortion builds gradually, but the signal doesn't clip harshly as it does with transistors. That is why musicians still use tube amps for guitars and bass. They can control distortion at the preamp. The best single input tube pre-amps for studio and stage can cost $1800 new. Good guitar amps are hundreds of dollars, and better ones cost thousands. Musicians treasure amps from the 1960s for their vintage sounds. They spend small fortunes getting them refurbished with new capacitors, controls, safety upgrades, and speaker cones.

In audiophile tube systems, the distortion is very low, but what is there can be important to the overall listening experience. It adds some of what the French call, "Je ne sais quoi". The idiom means "I don't know what (that is, but it adds something I like)."

I designed and built my first high quality tube amp mainly to test for myself whether they sounded “better” than solid state, or that was just a bit of audiophile lore. I was lucky enough, having worked for Tektronix and Bruel and Kjaer, to have access to the best quality instrumentation to characterize and optimize them. I noted, as you mentioned, that distortion when overdriven was very “gentle” with tubes compared to SS. My reference SS amp (a Crown) might have an IM of .05% at 100W, but >1% at 110W. But what I really noticed was the difference in the nature of the THD and IM distortion. The SS amp had more odd order distortion products (3rd, 5th harmonics), while the tube amp was mostly even order (2nd, 4th harmonics). My thought was that perhaps we are more accustomed to listening to listening to chords composed of even order harmonics in western music. Is that reason correct? No idea, but the measurements were, and I personally find the tube amps more comfortable to listen to, especially at high levels for long periods of time. BTW, one typical flaw of tube amps is hum and thermionic emission noise. I minimized this by using completely separate power supplies (zero AC on the amplifier chassis) - one for the filaments (DC) and one (very heavy) for the B+ and bias supplies. In both cases these were linear (not switchers) with multisection choke input (LCLC) filters. The result is that even at full gain, the hum and noise is absolutely inaudible (>90 dB down).

Yes, it seems to be that odd-even order difference in THD and IM that makes the difference. If there's a LOT of headroom in a solid state system, there's not much audible difference. At Davidson in the 1970s, we had a Crown DC-300A driving two Klipsch La Scala speakers (very large and very efficient) in a small bar/hall we used for intimate concerts. We could send people running out of the room with more than 10-15 Watts, so we had plenty of headroom. It sounded great and LOUD, even with the amp gain quite low. The console output/amp input was limited ever-so-slightly, to protect the speakers against mic drops and other undesirable noises that could damage speakers. Anything that hit +6db peak on the console triggered the limiter to clamp it down there. The meters on the amp barely twitched above -20db.

Your amp design sounds great! You nailed the issue with power supplies in tube amps. Most commercial producers cut corners on that and wind up with 15-20 db more noise. Is your pre-amp tube as well?

Bill, the preamp is purely a passive device to switch between inputs plus an infrared linked motorized volume control. The CD changers drive the amp directly - their 1V output will drive them to full power. Each amp has 4 KT88s and a pair of 6SN7s - one for drivers and one for preamp and phase inverter. Everything is directly coupled except the drivers to finals. The tubes are configured as parallel/push pull wired Ultralinear using a 280W output transformer from Hammond. I originally had them switchable so I could switch the screens to fixed bias, ultralinear or triode (screens tied to the control grid), but after extensive testing, decided on ultralinear. The plates are at 525 VDC and idle at 40 ma/tube. The amp will produce much higher output, but I rate it at 100W with <1% THD. That sounds awful compared to a solid state amp, but quite good for tubes. I also use an adjustable proprietary negative feedback design (always a topic of hot discussion with audiophiles), which is an absolute necessity to keep THD and IM reasonable even with the negative feedback from the ultralinear connection. I originally started building a Triode amp with gigantic 500W RF triodes, but the plate voltage and exposed plate caps were just too dangerous, so I reverted to a more conventional design. I spent a long time reading the patents of designers like McIntosh and a long time building really hum free power supplies. Since the speakers I drive are ultraefficient and able to handle all the power with ease, They, like your LaScalla/Crown setup will produce 110-120 DBA (measured) with only 10-20 watts, so LOTs of headroom for transients. Crown makes excellent amps - I used one for testing when I designed the speakers before designing the tube type amps. I’d you ever get the urge to build some big tube amps, I’ll be glad to share the schematic and parts list.

Cool! I might have to tap your expertise. 110 dbA is more than most people want for most practical purposes! My wife screams at me if I go over about 85 db! I then reluctantly enjoy the solace of a good pair of headphones.

Cheers,
Chris