Vincent Audio C-60 CD player Measurements

Sidebar 3: Measurements

I used Stereophile's loan sample of the top-of-the-line Audio Precision SYS2722 system to measure the Vincent C-60 (see www.ap.com and the January 2008 "As We See It"). For some tests, I also used my Audio Precision System One Dual Domain and the Miller Audio Research Jitter Analyzer.

The Vincent's output preserved absolute polarity (ie, was non-inverting) from both its balanced and unbalanced jacks, with and without the FET buffer. The XLR jacks are thus wired with pin 2 hot. The remote control offers volume up and down buttons, with the maximum level indicated as "26" on the C-60's front-panel display. This control operated in accurate steps of 1dB; a setting of "20" was 6dB below full level. The maximum output at 1kHz from the tubed outputs, with the volume control set to "26," was 5.1V balanced and 2.53V unbalanced. Switching in the FET buffer stage raised these levels to 5.3V and 2.63V, respectively, owing to the lower output impedance of this stage. The maximum unbalanced output was a very audible 0.5dB higher than the CD standard's 2V, which will be audible in direct comparisons with other players.

The FET stage's output impedance was 198 ohms at all frequencies from the balanced XLRs, 99 ohms at all frequencies from the unbalanced RCAs. By contrast, the balanced tubed output impedance measured 660 ohms at high and middle frequencies, increasing into 3k ohms at 20Hz. The unbalanced output impedance was half these figures, suggesting that the balanced output is a true balanced topology, with both phases of the signal actively driven.

Brian Damkroger noted that "The C-60 eventually did play every disc I threw at it, but something about its drive or error-correction circuitry was finicky." Though the player seemed a little fussy about how a disc was placed in its disc well, I found that its error correction, assessed with the Pierre Verany Test CD, was superb. The C-60 didn't suffer from audible glitches in its output until the gaps in the test disc's data spiral reached 2.4mm in length. However, monitoring the error flag in the player's S/PDIF output with RME's DIGICheck program revealed that errors were being interpolated once the gaps in the data spiral reached 2mm.

The Vincent's frequency response rose very slightly in the top octave (fig.1, top pair of traces), but with preemphasized data there was a prominent suckout in the treble (fig.1, bottom traces). Channel separation was excellent, at better than 90dB at 1kHz.

Fig.1 Vincent Audio C-60, frequency response at –12dBFS into 100k ohms (left channel blue, right red) and with preemphasized data (left cyan, right magenta; 0.25dB/vertical div.).

Analyzing the spectrum of the Vincent's balanced tubed output while it decoded data representing a dithered 1kHz tone at 90dBFS (fig.2, 1/3-octave analysis; fig.3, FFT analysis) revealed minimal level error and a noise floor dominated by the dither noise used to encode the signal. There was a commendable absence of both harmonic and supply-related spuriae, though at low frequencies the right channel was very slightly noisier than the left. Linearity error was very low to below 100dBFS, and again was dominated by the recorded dither noise below that level. With its excellent linearity and low noise, the C-60's reproduction of an undithered tone at exactly 90.31dBFS was superb, with the three DC voltage levels defined by this signal clearly evident and with excellent waveform symmetry (fig.4).

Fig.2 Vincent Audio C-60, 1/3-octave spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with 16-bit data (right channel dashed).

Fig.3 Vincent Audio C-60, FFT-derived spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with 16-bit data (left channel blue, right red).

Fig.4 Vincent Audio C-60, waveform of undithered 1kHz sinewave at –90.31dBFS, 16-bit data (left channel blue, right red).

I wasn't surprised to find that the C-60's tubed output suffered from rather high distortion. Fig.5, for example, shows the spectrum of the player's balanced output with a full-scale signal into the benign 100k ohm load with the volume control set to its maximum ("26"). A regular series of distortion harmonics is evident, with the second and third harmonics around –70dB in the left channel (0.03%, blue trace), but the second harmonic highest in the right channel at –64dB (0.06%, red trace). Halving the level by lowering the volume control to "20" (fig.6) dropped the level of the distortion harmonics, though the level of the second harmonic in the right channel was still –69dB (0.03%).

Fig.5 Vincent Audio C-60, tubed output, volume control at "26," spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

Fig.6 Vincent Audio C-60, tubed output, volume control at "20," spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

I was surprised to find that switching in the FET buffer did not change the distortion's level or spectrum (fig.7). Obviously, the FET stage follows the tubed stage with the latter's bent transfer function. So while the FET stage won't change the C-60's intrinsic sound quality, it does enable the player to better drive preamplifiers with low input impedances. Without the FET stage, the distortion at low frequencies is high, even at low levels (fig.8). With the FET stage in-circuit, the low-frequency distortion remains at acceptable levels (not shown); I was not surprised that BD found that "The solid-state output stage . . . had a little more bottom-end punch than the tubed stage, and notes stopped and started with a bit more precision and authority with the transistors."

Fig.7 Vincent Audio C-60, FET output, volume control at "20," spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

Fig.8 Vincent Audio C-60, tubed output, volume control at "09," spectrum of 50Hz sinewave, DC–1kHz, at 0dBFS into 600 ohms (left channel blue, right red; linear frequency scale).

Even from its tubed output, the Vincent C-60 performed well on the demanding high-frequency intermodulation test into high impedances (fig.9): the 1kHz difference tone resulting from an equal mix of 19 and 20kHz tones lay at –69dB (0.03%). Again, this was not affected by the FET buffer stage.

Fig.9 Vincent Audio C-60, tubed output, volume control at "26," HF intermodulation spectrum, DC–24kHz, 19+20kHz at 0dBFS into 100k ohms (linear frequency scale).

Finally, the C-60 offered superb jitter rejection. While sidebands can be seen in the spectrum of its output while it played the 16-bit Miller/Dunn J-Test tone (fig.10), these are actually at the residual level of the harmonics of the low-frequency, LSB-level squarewave. These are not being exaggerated by the C-60, nor are any other sidebands visible, and the central peak that represents the high-level 11.025kHz tone is narrow and free from low-frequency skirts.

Fig.10 Vincent Audio C-60, tubed output, volume control at "26," high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: CD data (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

While its digital performance is excellent, the Vincent C-60's analog behavior is dominated by the bent transfer function of its tubed circuits, even when the solid-state buffer circuit is engaged. The C-60's tubed output does need to be used with a preamplifier offering a high input impedance. The C-60 shouldn't have had problems with either of the two preamplifiers used by BD: the Sutherland Direct Line Stage, which offers an input impedance of 40k ohms over most of the audioband, dropping to 20.5k ohms at 20kHz; and the Placette Active Line Stage, whose input impedance is 18.5k ohms at all frequencies. The mystery to me is why, if the FET-buffered output preserves the sonic signature of the tubed stage, BD found it to sound significantly different from the tubed output.—John Atkinson

COMPANY INFO
Vincent Audio
US distributor: WS Distributing
3427 Kraft SE
Grand Rapids, MI 49512
(616) 885-9809
ARTICLE CONTENTS
Share | |
COMMENTS
WillWeber's picture

Hello Brian,

Well I’m confused.

You state: “Should an audio component accurately reproduce the signal it's fed, or should it evoke the sound and feel of live music? Accuracy or musicality?”

You suggest that traditionally solid state has claimed to be  “more accurate” (by measurements) while tubes sounded more “musical.” But then you state: “…both camps have eliminated the obvious colorations of their respective technologies, and the levels of performance of today's best tubed and solid-state gear have converged.”

So if these “obvious colorations” are eliminated and the sound quality has “converged” over the years, doesn’t that imply both have become more accurate? (Colorations are an inaccuracy right?) Then where does "Musical" lie in this new terrain?

Then this new CD player has a choice of tube and solid state paths. And you state that is has: ”…a recognizable sonic signature—actually, more than one,…” So is this not a throwback in technology? You even detail how the tube path is sweet and warm (“…a little sweeter and more golden than reality”),  while the solid state path “…more accurate and more tonally neutral...”

So what am I missing here?

Thanks in advance for any helpful explanation,

WillW

PS: Opening Panda’s Box - Tube vs SS. Tubes are sweeter and warmer, my experience too, overall. But I find this sound syrupy and not so accurate. I find (high end) SS more like a live performance. In fact, I can make my synth sound sweeter, more like tubes, by adding a touch of 2nd and 3rd harmonics. That experimental data point suggests that this might be the allure of tubes, for those who prefer a more saccharine sound, albeit less accurate. (OK, all you tube guys get out your guns!) "Musical" might be a subjective term to some, but live performance is musical to me, and tubes (traditionally) don't have as much live character, again to me.

Brian Damkroger's picture

Hi WIll. Thanks for the comment.

First, there's the question of semantics, and how we agree to use the words accurate and musical.  I tried to use the former to refer to the claim made by early solid-state proponents, which was that their gear had a measurably more accurate reproduction of the input waveform. The gear also sounded hard, cold, hazy, and any number of other not so great adjectives.  I used musical, on the other hand, to describe a classic tube sound - harmonically rich but with all the other baggage that plagued the early tube gear.  I didn't use accurate, or musical for that matter, to mean accurately reproducing the original musical event.  In my vernacular, neither accurate nor musical were particularly in this latter regard.

Over time, both tube and solid-state gear dramatically reduced their most obvious audible colorations, both sounding more like music and not coincidentally, more like each other.  Both did, as you note, grow to more nearly approach, or more accurately reproduce, the sound of the original recorded event.  Accurate and musical, as I originally used the terms, became less applicable and less relevant.  

You're right, the Vincent gear is a bit of a throwback in that it offers the two different output stages.

Thanks again for the comment.

Brian

WillWeber's picture

Hi Brian,

Thanks for your clarification, that helps me understand your position much better.

Semantics are often a struggle in communications, as the forums attest to (quite strongly!). To me, as a scientist, "accuracy" has a very specific meaning that is carefully defined to the nth degree in the metrological context. I duly note that you refer to the context of early SS measurements, which were certainly incomplete, dubious even. In fact, these types of measurements have improved, but still I think they are lacking. There is a tendency for the disease that I call mononumerosis, that is trying to present data one dimensionally to allow easy comparisons. Meaningful measurements are much more complicated. And there are undoubtedly parameters that matter but are not measured, or at least not measured well enough.

I agree that many of the early (as well as cheap contemporary) SS designs are harsh and cold sounding, and would not seem to be so accurate, or musical. Some of the modern ‘phile designs though are silky smooth, dynamic, and detailed, and actually measure quite well too. I suspect that manufacturers are using more thorough and accurate measurements to assist in their R&D, which has enabled such progress, both valve and SS gear. Tubes still have more distortion, though of the kinder “musical” low harmonics, and output stages always require that unfortunate tranny output for impedance matching--with the consequential low Q and speaker impedance interaction. I hear these compromises as coloration and lack of control and dynamics (with most speakers). But I do think that good tube designs can be more useful in pre stages and line buffers, just not my preference (yet). There remain the issues of short life, excess heat, warm-up time, energy consumption, and microphonic behavior too.

My musician side, with context of “accuracy” and “musical” being as you describe--reproducing the live event as realistically as possible--still prefers high end SS over the sweetened sound of tubes, which seems to be largely unavoidable.

I appreciate your taking the time to explain your usage of these terms. (Aren’t you busy at the show in Germany?)

Cheers,

Will

X
Enter your Stereophile.com username.
Enter the password that accompanies your username.
Loading