Halide Design DAC HD D/A converter Measurements

Sidebar 3: Measurements

I measured the Halide DAC HD with Stereophile's loan sample of the top-of-the-line Audio Precision SYS2722 system (see www.ap.com and the January 2008 "As We See It"); for some tests, I also used my vintage Audio Precision System One Dual Domain.

I connected the Halide to one of the USB ports on my Intel MacBook; the Mac USB Prober utility reported the DAC HD as being the "Halide Design DAC HD" manufactured by "Halide Design," and listed the serial number as "(C) 2010 Wavelength Audio, ltd." The last refers to Gordon Rankin's Streamlength proprietary asynchronous operating code for the TAS1020B USB receiver chip, and USB Prober confirmed that the DAC HD does indeed operate in the preferable isochronous asynchronous mode, with 24-bit word length and data sampled at 44.1, 48, 88.2, and 96kHz.

The DAC HD's maximum output level at 1kHz was 2.035V, and its output preserved absolute polarity (ie, was non-inverting). The output impedance was a low 199 ohms at all frequencies. Halide says that the DAC HD's digital filter is optimized for "a much more natural sound than the standard 'fast roll-off' interpolation filter more commonly used in digital audio." Fig.1 shows the DAC HD's impulse response; symmetrical in the time domain, it has just two cycles of ringing before and after the impulse. As usual with this kind of filter, there is a slight rolloff just below half the sample rate. With 44.1kHz data (fig.2, cyan and magenta traces), the output is down by 2dB at 20kHz, which will be inaudible. With 96kHz data (fig.2, blue and red traces), the output rolls off smoothly above the audioband, reaching –2dB at 30kHz and –6dB at 40kHz. Channel separation (not shown) was superb, at >100dB below 2kHz and still 78dB at the top of the audioband.

Fig.1 Halide Design DAC HD, impulse response (4ms time window).

Fig.2 Halide Design DAC HD, frequency response at –12dBFS into 100k ohms with data sampled at: 44.1kHz (left channel cyan, right magenta), 96kHz (left blue, right red) (1dB/vertical div.).

Linearity error (not shown) was negligible down to below –110dBFS, and there was a complete absence of modulation noise (fig.3). Spectral analysis of a dithered 16-bit/1kHz tone at –90dBFS revealed an absence of harmonic or supply-related spuriae (fig.4, top pair of traces, fig.5, cyan and magenta traces). Increasing the bit depth to 24 dropped the noise floor by about 8dB (fig.4, bottom traces, fig.5, blue and red traces), suggesting ultimate resolution of between 17 and 18 bits, which is good when you consider that the DAC HD has to get its power from the host computer's USB bus. The Halide's reproduction of an undithered 16-bit tone at exactly –90.31dBFS (fig.6) was a little noisier than the best DACs, but the three DC voltage levels described by these data can be resolved, and very little DC offset is present. Increasing the input word length to 24 bits gave a good if noisy representation of a sinewave (fig.7).

Fig.3 Halide Design DAC HD, spectrum of 1kHz sinewave, DC–1kHz, at: 0dBFS (left channel cyan, right magenta), –90dBFS (left blue, right red) (24-bit data, linear frequency scale).

Fig.4 Halide Design DAC HD, 1/3-octave spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS, with 16-bit data (top) and 24-bit data (bottom) (right channel dashed).

Fig.5 Halide Design DAC HD, FFT-derived spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS, with: 16-bit data (left channel cyan, right magenta), 24-bit data (left blue, right red).

Fig.6 Halide Design DAC HD, waveform of undithered 1kHz sinewave at –90.31dBFS, 16-bit data (left channel blue, right red).

Fig.7 Halide Design DAC HD, waveform of undithered 1kHz sinewave at –90.31dBFS, 24-bit data (left channel blue, right red).

As it is powered from the USB bus, the Halide DAC HD was just starting to clip with a 0dBFS tone, the second and third harmonics lying at –64dB (0.06%, not shown). But the DAC HD offered very low levels of harmonic distortion at levels below full scale. Fig.8, for example, was taken with the Halide decoding a 50Hz signal at –10dBFS into 100k ohms; the only harmonic visible is the second, but this lies more than 100dB down (<0.001%). With its slow-rolloff, time-domain–optimized reconstruction filter, the DAC HD's performance on the high-level, high-frequency intermodulation test is compromised by the presence of ultrasonic image energy folding back into the audioband (fig.9). However, as all the audioband aliasing products lie at or below –100dB, they should have no audible consequences. Actual intermodulation distortion is low, the difference product at 1kHz lying at –94dB (0.002%).

Fig.8 Halide Design DAC HD, spectrum of 50Hz sinewave, DC–1kHz, at –10dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

Fig.9 Halide Design DAC HD, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 0dBFS into 100k ohms (left channel blue, right red; linear frequency scale).

Finally, because it operates the USB link in the asynchronous mode, in which the DAC and not the computer controls the clocking of the data, the DAC HD very effectively rejects word-clock jitter. Fig.10 shows the narrowband spectrum of the Halide's output while it decodes 24-bit J-Test data. No jitter-related sidebands at all are visible, and notably, there is very little spreading of the spectral spike representing the 11.025kHz tone.

Fig.10 Halide Design DAC HD, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz: 24-bit data via USB from MacBook (left channel blue, right red). Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz.

Despite its small size and equally small price, Halide Design's DAC HD offers superb digital audio engineering.—John Atkinson

Halide Design
(858) 224-3551
Share | |

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