Wadia 861 CD player Measurements part 2
Fig.7 Wadia 861, spectrum of 50Hz sinewave, DC-1kHz, at 0dBFS into 100k ohms (linear frequency scale).
Fig.8 Wadia 861, Filter A, HF intermodulation spectrum, DC-25kHz, 19+20kHz at 0dBFS into 600 ohms (linear frequency scale).
Finally, I examined the Wadia 861's rejection of word-clock jitter using the Miller Audio Research Jitter Analyzer and a CD-R carrying the analytical signal, this a high-level Fs/4 (11.025kHz) sinewave over which is overlaid the LSB toggling on and off at a low frequency (229Hz). (It is important to note that, contrary to what you may have read on the Usenet newsgroups, this signal must be undithered to maximally excite wordclock jitter in the DAC under test.)
The solid trace in fig.9 shows a high-resolution spectrum of the Wadia's unbalanced analog output while it decodes data fed to its digital input by a PS Audio Lambda CD transport via a 6' electrical S/PDIF link. (ClockLink was turned off for this measurement.) The absolute jitter level is a very low 190 picoseconds, and the main jitter components are the pairs of sidebands at ±229Hz—clearly data-related and indicated with red "4" numerical markers—and ±16Hz (purple "1," circled in magenta).
Fig.9 Wadia 861, high-resolution jitter spectrum of analog output signal, PS Audio Lambda 2 transport via 6' Apature S/PDIF link (11.025kHz at -6dBFS with LSB toggled at 229Hz). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz. Grayed-out trace is similar analysis for internal CD playback.
Repeating the test using the 861's own drive actually increased the jitter to 191.6ps, which is still low in absolute terms. This surprised me, considering that the internal transport benefits from Wadia's ClockLink topology, which should reduce the DAC's susceptibility to word-clock jitter. I turned the front-panel display off and the measured jitter level dropped to 176.4ps. It appears that there is some interference between the display and the D/A circuitry with the internal CD transport but not with external digital sources; while the measured jitter levels are low in absolute terms, there is still some noise-floor contamination, as can be seen from the red-circled spikes in the grayed-out trace in fig.9.—John Atkinson