Page 6 of 6
A major issue with this audio board is that it’s very dependent upon the quality of the power supply used to drive it. A noisy 5V switch-mode power supply unit (PSU) resulted in 76dB dynamic range from Line Out, but a quiet one (Apple phone charger) gave an excellent 110dB – an improvement of 34dB gained solely by using a better power supply.
More surprisingly, this also cleaned up the digital S/PDIF output to a Chord 2Qute external DAC, allowing it deliver the full 121dB dynamic range of which it is capable.
A good supply would be a battery of some sort, lead-acid or lithium ion, for top audio results, or a smoothed linear supply, bearing in mind that 5V is needed and batteries deliver 6V. Switch-mode supplies generate rubbish and the audio card is sensitive to this (the Apple charger is switch mode, but clean). See p320 of the WM5102 datasheet for more on this.
With a full level 0dB digital playback signal the audio board gave exactly 1V out maximum from HeadSet (black socket) before overload. However, it overloads after the volume control and with a 0dB signal volume had to be reduced a long way to avoid overload and the distortion it generates. Overload such as this would be obvious on headphones and only with insensitive ‘phones would more than 1V be needed so there are no issues here.
The Cirrus Logic audio board with loudspeaker leads soldered in and a
suitable Maplins 5V, 2A smoothed power supply connected.
With a -60dB distortion test signal, gain (volume) could be set to maximum and under this condition output measured 37mV, indicating the audio board needs to swing 3.7V rms to deliver full output without overload – not possible from a 5V d.c. power supply without using power line d.c. voltage convertors. At full gain the board’s DAC was able to deliver 0.05% distortion from a 24bit resolution signal – a surprisingly good result, up with that achieved from hi-fi convertors. EIAJ Dynamic Range measured 113dB – also a very good figure, well above CD at 102dB, if not quite up with the hi-fi average of 115dB or top DACs at 122dB. It shows the board’s DACs offer good results, however.
With a 192kHz sample rate file, frequency response measured flat to 56kHz (-1dB) our analysis shows, at half volume. Surprisingly, the volume control curtails this a bit at high settings, bringing the upper limit down to 20kHz, but this isn’t going to be easily audible nor a real life setting. So the board exploits 192 hi-res digital well enough.
Line Out (green socket) offers a fixed output unaffected by the volume control. Maximum level has been set to 1V to avoid output overload, distortion (24bit) measuring 0.07% at -60dB and EIAJ Dynamic Range 110dB – similar results to HeadSet with volume reduced to give 1V out, so Line Out does not offer better quality, only fixed volume.
The digital S/PDIF output was connected to a Chord 2Cute DAC to see if its intrinsic 121dB dynamic range value could be exploited to achieve better quality than the board’s internal DACs. Interestingly, power supply noise made a massive difference here, dynamic range improving from 106dB with a noisy PSU, to a full 121dB through the 2Cute with a good PSU. So using an external DAC connected to S/PDIF out does offer better quality than the internal DACs. However, only DACs from Chord, ESS (Sabre32 9018) and Burr Brown (PCM1765) improve on the audio board’s DACs that, in themselves, offer relatively good results. The volume slider of the LXMusic player, even though it would appear to be digital, does not affect S/PDIF out.
Recording via the Line In (pink socket) showed a 1V input was needed for a full level 0dB signal as standard default. However, record gain can be increased by those with the time and inclination to delve into writing a suitable command line script. Recordings at 16/44.1kHz delivered similar quality to CD, with bandwidth to 20kHz and 92dB dynamic range. Our board would only record at 16bit resolution, not 24bit and the WM5102 data sheet alludes to Normal and High Performance modes, but which mode the board uses is unstated. The ADCs are 24bit it seems from the datasheet.
The loudspeaker output delivered 3V at 3% distortion and above this output overload (clipping) set in. Frequency response reached 90kHz however, with a 192kHz sample rate signal. We did not use an output LP filter, but one is required the datasheet says (p318) to remove all the switching noise. There was no d.c. offset, differential or to ground.
The Cirrus Logic audio board delivered very good results from its on-board DACs, reaching full hi-fi spec. The ADCs worked well too, offering CD quality. Only the small loudspeaker power amplifiers were a bit limited, but that is to be expected considering they are built into a tiny chip and an external low pass filter (22µH/3µF) is needed to clean the output. NK
Frequency response (-1dB)
Separation (1kHz) 110dB
Noise (IEC A) 110dB
Dynamic range 113dB