Einstein Audio Components The Tube Mk.II preamplifier Measurements

Sidebar 3: Measurements

I used Stereophile's loan sample of the top-of-the-line Audio Precision SYS2722 system (see the January 2008 "As We See It" and www.ap.com) to examine the Einstein The Tube Mk.II's measured behavior; for some tests, I also used my vintage Audio Precision System One Dual Domain. Not having seen Michael's review when I performed the measurements, I used the CD input with the stock tubes. (All the inputs are nominally identical.)

Assessed at its balanced outputs, the Einstein offered the same maximum gain—a sensibly moderate 12dB—for both balanced and single-ended input signals. The unity-gain setting of the volume control was 11:00 and the control affected only the level of the balanced outputs. (The single-ended outputs appeared to be pass-throughs from the single-ended inputs.) The Einstein preserved absolute polarity for both balanced and single-ended inputs, the XLRs wired with pin 2 hot. The balanced input impedance was usefully high at 90k ohms across the audioband, but the unbalanced input impedance was lower, at 14k ohms.

Though the balanced output impedance was 95 ohms at all frequencies, this is almost twice the specified value of 50 ohms, which suggests that the specification refers to the impedance of each polarity; ie, half the total. This is still very low for a tubed design. As a result, there is no difference between the preamplifier's responses when it drives 100k ohms (fig.1, blue and red traces) and 600 ohms (cyan, magenta). These traces were taken with the volume control at its maximum setting; the high frequencies start to roll off above 20kHz, reaching –3dB at 180kHz. Einstein's The Tube Mk.II has a very wide bandwidth—with the volume control set to unity gain, the response was down just 0.4dB at 200kHz (fig.1, green and gray traces) and the bass was flat to 20Hz, even into 600 ohms, which suggests that the design is direct-coupled. However, there was a 0.25dB channel imbalance at all settings of the volume control.

Fig.1 Einstein The Tube Mk.II, frequency response at 1V into 100k ohms (left channel blue, right red) and 600 ohms (left cyan, right magenta) with volume control at maximum, and at 1V into 100k ohms (left green, right gray) with volume control at unity gain (11:00). (0.25dB/vertical div.)

Channel separation (not shown) was better than 100dB in both directions below 2kHz, but rose above that frequency to a still-good 80dB at 30kHz. The Tube Mk.II was superbly quiet: With the balanced input shorted but the volume control at its maximum, the unweighted wideband signal/noise ratio (ref. 1V) was 83.5dB. Restricting the measurement bandwidth to the audioband improved this figure to 88dB, while switching an A-weighting filter into circuit gave 90.75dB!

Even with so little noise, the Einstein's THD remained buried under the noise floor for output levels below a few hundred millivolts, as revealed by the downward slope of the traces at the left of fig.2—a fixed amount of noise becomes an increasingly smaller percentage of the signal level as the signal increases. Actual distortion starts to rise above 1V into 100k ohms and above 500mV into 600 ohms, but is still low in absolute terms by 4V, which is about the maximum the preamp will be asked to deliver in real-world systems. The Tube Mk.II clips (defined as 1% THD) at a high 22V into 100k ohms and manages a high 15V into 600 ohms. This preamp offers an unusually high dynamic range. Distortion is higher for an unbalanced input signal (fig.3), reaching 0.1% at 1.1V into 100k ohms.

Fig.2 Einstein The Tube Mk.II, THD+N (%) vs balanced output voltage at 1kHz into (from bottom to top): 100k, 600 ohms.

Fig.3 Einstein The Tube Mk.II, THD+N (%) vs unbalanced output voltage at 1kHz into (from bottom to top): 100k, 600 ohms.

Fig.2 was taken from the left channel, but the right channel was actually more linear, as can be seen in fig.4, which plots the THD+noise percentage against frequency at 2V into 100k and 600 ohms. Into 600 ohms, the right channel (magenta trace) didn't offer much more distortion than the left channel did driving 100k ohms (blue). I suspect that both the channel gain and the ultimate linearity will depend on the specific tubes used.

Fig.4 Einstein The Tube Mk.II, balanced output, THD+N (%) vs frequency at 2V into: 100k ohms (left channel blue, right red), 600 ohms (left cyan, right magenta).

In fully balanced mode, the Einstein's distortion signature was predominantly the subjectively innocuous second and third harmonics (fig.5), with a greater degree of second harmonic in the left channel (blue trace) indicating that the tubes used in the hot and cold signal phases were not as well matched in that channel as they were in the right channel (red). Dropping the load to 600 ohms increased the levels of these harmonics by about 10dB but preserved their relationship to one another (not shown). However, driving the preamp with an unbalanced signal for the same output level raised the second harmonic to –60dB (0.1%) left and –64dB (0.07%) right (fig.6). This will not be audible, but suggests that the preamplifier's design is optimized for fully balanced operation. Note, by the way, that while the right channel's noise floor in these two graphs (red traces) is generally clean, the left channel (blue traces) has low-level spuriae present at the 60Hz AC power-line frequency and its harmonics. Again, this will have no audible consequences, but it does suggest that the left channel's circuit-board layout is not as optimal as the right's.

Fig.5 Einstein The Tube Mk.II, balanced output, spectrum of 1kHz sinewave, DC–10kHz, at 2V into 100k ohms (left channel blue, right red; linear frequency scale).

Fig.6 Einstein The Tube Mk.II, balanced output but single-ended input, spectrum of 1kHz sinewave, DC–10kHz, at 2V into 100k ohms (left channel blue, right red; linear frequency scale).

Finally, fig.7 shows the spectrum of the Einstein's output while it drove a high-level mix of 19 and 20kHz tones into the demanding 600 ohm load. The 1kHz difference product lies at –72dB in the left channel (0.028%) and at –76dB in the right (0.015%), which implies good high-frequency linearity even with the output stage having to deliver significant current.

Fig.7 Einstein The Tube Mk.II, balanced output, HF intermodulation spectrum, DC–24kHz, 19+20kHz at 2V peak into 600 ohms (linear frequency scale).

It may use tubes, but Einstein Audio's The Tube Mk.II is an impressively well-engineered preamplifier that offers a wide bandwidth, high dynamic range, and low levels of noise and distortion. It will also drive low impedances with aplomb.—John Atkinson

Einstein Audio Components
US distributor: Aaudio Imports
4871 Raintree Drive
Parker, CO 80134
(720) 851-2525