Musical Fidelity X-10D line-level preamplifier Measurements

Sidebar 3: Measurements

Although it could be conjectured that the presence of tubes in the X-10D might add a certain thermionic something to its sound, there was nothing in the measurements that would confirm that hypothesis. In fact, the Musical Fidelity appears to use a well-engineered circuit, with no compromises that could be laid at the feet of its tubes.

The A-weighted S/N Ratio, for example, was an excellent 95dB (ref. 1V output), though this did decrease to 68.4dB, unweighted, across a wide 10Hz-500kHz bandwidth. The buffer's input impedance was high, at least 320k ohms, while its output impedance was a uniform 26 ohms across the audio band, usefully lower than specified. It didn't invert polarity. As Muse Kastanovich noted, however, the Musical Fidelity does have a small amount of gain: 0.98dB. Despite MK's protestations, I feel this is large enough to cast doubt on any comments made about the unit's sound quality without this gain being compensated for. The X-10D won't necessarily sound louder, but, all things being equal, it might well sound more present, more detailed, more dynamic. I note with interest that MK's comments actually were in the opposite direction: that the Musical Fidelity tended to make the sound less involving and less dynamic.

The X-10D's frequency response, plotted out to 200kHz in fig.1, two octaves higher than is our usual practice, is flat within the audio band and doesn't reach the specified -0.2dB until 150kHz. The lower pair of traces in fig.1 is with the load impedance reduced from the standard 100k ohms to a punishing 1k. Other than a drop in level of 0.2dB, there is no change in response. Note also the excellent channel balance. The plot of crosstalk against frequency (fig.2) shows decreasing channel separation at low frequencies, this usually an indicator of increasing power-supply impedance. Separation is still good in the midrange and above, however.

Fig.1 Musical Fidelity X-10D, frequency response at 1V output into 100k ohms (top) and 1k ohms (bottom) (right channel dashed, 0.5dB/vertical div.).

Fig.2 Musical Fidelity X-10D, crosstalk (from bottom to top at 10kHz): R-L, L-R, unbalanced (5dB/vertical div.).

I haven't shown the plot of the X-10D's THD+N against frequency; at 2V into either 100k or 1k ohms, it's a straight line at the 0.015% level. This excellent linearity was confirmed by looking at a spectral analysis of the unit's output driving 50Hz at 2V into 1k ohms (fig.3). Even under this worst-case condition, the only significant harmonics that can be seen are the second- and third at -86dB (0.0005%). It's hard to believe that the circuit features tubes!

Fig.3 Musical Fidelity X-10D, spectrum of 50Hz sinewave, DC-1kHz, at 2V into 1k ohms (linear frequency scale). Note that the second and third harmonics are the highest in level at -86dB (0.005%).

Finally, fig.4 shows the THD+N percentage plotted against output voltage into both 100k and 1k ohms. The traces overlap, so immune is the X-10D to the load impedance. The drop in THD+N from 0.1V to 2.2V is due to the noise component becoming less and significant as a fraction of the measured level. But above the minimum distortion level at 2.2V, the X-10D rapidly becomes nonlinear, clipping (defined as 1% THD+N) at 7V RMS. This is much higher than the output of any current CD player I know of, though some Theta and Kinergetics models will overload the Musical Fidelity. These components do have powerful buffer stages of their own, however, making the use of an outboard stage unnecessary.

Fig.4 Musical Fidelity X-10D, distortion (%) vs output voltage into 100k ohms and 1k ohms .

Other than the subjectively misleading gain and the high crosstalk at low frequencies, this is excellent measured performance.—John Atkinson

Musical Fidelity
US distributor:Kevro International, Inc.
902 McKay Road, Suite 4
Pickering, Ontario L1W 3X8, Canada
(905) 428-2800