Soulnote D-2 Digital to Analogue Converter
The D-2 is Soulnote’s Digital to Analogue Converter from their middle range of products – Series 2
Step up from Series 1 to Series 2 and the first thing you will notice is the larger casework, the D-2 stands at 160mm tall, 5cm taller than the D-1N
Series 1 models all have the front panel milled from aluminium, step up to Series 2 and that continues around the sides, the build quality is first rate and the D-2 weighs in at 17kg. The top plate of the D-2 ‘floats’ on the chassis, it is not firmly attached or screwed down. Soulnote have always argued that the case construction is just as important the to overall performance as the actual components that make up the product itself.
The D-2 offers 5 digital inputs with USB-B, 2 x AES/EBU and 2 x RCA. There are both RCA and XLR analogue outputs, there is also a word clock input – Soulnote make a Clock Generator, called the Z-3 which serves as a great upgrade for the D-2
There is also switches on the rear to choose between FIR oversampling and a non-oversampling mode (PCM only), there is also a mono/stereo switch on the rear as well, you can double up with 2 and run a pair in dual mono mode.
In an industry first the D-2 utilises 4 ESS ES9038Pro chips in its design (2 per channel) to ensure it delivers the best possible performance.
When partnered with a Melco the presentation is vivid and vibrant, but there is a real sweetness and allure to the music, you get a wonderful sense of air and space around everything and it seems to make everything sound real and almost tactile when are sat listening to it, partnered with A-2 it makes for a seriously strong combination.
The DSS uses the LMX2594 from Texas Instruments
D/A converter formats that recreate the required master clock from standard clocks typically use a DDS (Direct Digital Synthesizer). The performance of external clock generators in this format guarantees extreme frequency accuracy (e.g. how many seconds of deviance will occur over tens of thousands of years), but jitter (fluctuations in waveform width arising in the clock) and the resulting phase noise are highly dependent on the performance of the DDS. By convention, audio equipment generally uses DDS-specific ICs with 10 ps (picosecond) jitter functions, but the D-2 is the first device to use Texas Instrument’s PLLatinum™ RF LMX2594 synthesizer for an audio DDS. The LMX2594 synthesizer was developed for radars and measurement instruments and boasts the world’s highest specifications with an efficiency index of -236 dBc/Hz and jitter of 45 fs (femtosecond, 1/1000 of a picosecond). This ultralow jitter DDS results in an overwhelmingly clear and deep spatial presentation when inputting the high-precision external clock and using the built-in TCXO.
Supports 10 MHZ external standard clock inputs
Using a DDS with jitter functionality that surpasses quartz means that the D-2 comes with 10 MHz external standard clock inputs, a SOULNOTE first. The internal clock relies on a high performance TCXO boasting stunning phase noise (-121 dB at 10 Hz offset). It also features a high frequency relay with mechanical contacts to avoid additional jitter from the IC when switching between the internal clock and an external clock generator. The internal clock power supply shuts off and vibrations stop when using an external clock generator, this way removing any risk of mutual interference.
Uses 4 ES9038PRO (an industry first)
2 ES9038PRO circuits are employed in each channel for the upper and lower differential complimentary inputs in this SOULNOTE original fully symmetric non-NFB discrete amplifier, completely driving the initial stage in the non-NFB differential amplifier. The IV circuits are extremely simple IV circuits with only one resistor to eliminate TIM distortion that can be so problematic with feedback type IVs. With an extremely powerful 120 mA, the two ES9038PROs are used in each channel to allow natural music playback bursting with life. The ES9038PRO operates in synchronous mode to obtain the highest sound quality.
In addition to the conventional FIR oversampling mode, the D-2 also has the new non-over sampling mode, which is basically a non-NFB version adapted to operate in the digital domain. This mode does not generate the pre- and post-echoes observed in FIR oversampling. It also boasts extremely low noise levels with analog outputs, with no difference in levels between the two modes.
Impulsausgangswellenform mit einem FIR-Oversampling-Filter (D-2-Analogausgang)
Pre- and post-echoes are artificial “sounds” created by data calculations before and after the data is interpolated. They make waveforms like sine waves appear smoother, but calculation algorithms can affect the sound quality or ambiguity and be added to the time axis.
Impulse output waveform in Non-Over Sampling mode (D-2 analog output)
This waveform can only be achieved by combining a non-NFB discrete amplifier with superb transient response performance. Because music waveforms are actually a series of impulse waveforms of different heights, the non-over sampling mode eliminates any ambiguity in the time axis information, creating more realistic and natural sound quality and an expansive listening environment to the human ear, which is extremely sensitive to the time axis.
* The non-over sampling mode cannot be used with DSD.
Dual mono construction using fully discrete non-NFB differential amplifiers
The SOULNOTE original discrete fully symmetric non-NFB differential amplifiers are arranged in a dual mono construction that fully isolates each channel, including the power supply rectifier. Connecting cables have been eliminated for sound signals and in the power supplies. Wiring from the transformer has also been shortened, since this is a major cause of current loops.
400 VA bifilar wound high regulation toroidal transformers with eight windings on the secondary side
The D-2 incorporates the newly developed power transformers with eight windings on the secondary side that are capable of regulation similar to that of very high power amplifiers. Independent rectifier circuits in each stage prevent interference. Because transformers are a source of vibration, SOULNOTE has used only one in each unit to simplify the vibration type and has constructed the unit body so that any unwanted vibrations are transmitted away from the body through a spike placed directly under each transformer.
Noise countermeasures that don’t rely on filters
The D-2 uses 4 boards wrapped in 70 micron thick copper foil that eliminates noise by optimizing the printed circuit board layout. Noise countermeasures that don’t rely on filters that can impact sound quality like flight, etc. are employed throughout, and each mode keeps unwanted radiation below the noise floor.
An independent digital board layout with consideration of future upgrades
The digital boards are independent structures with one portion connected to the rear panel, allowing for upgrades in the future. Analog circuit connections use soldered pin heads, eliminating sound quality deterioration caused by contact resistance in connectors
Expandable to dual mono DAC
You can select between STEREO, MONO Lch, and MONO Rch modes on the D-2. In MONO mode, the ES9038PROs in the opposite channel are disabled, thus doubling the power margin and making channel separation practically infinite.
Uses the groundbreaking Bulk Pet transfer method
USB-Audio generally transfers data using the Isochronous transfer method. The D-2 uses Bulk Pet, a new Bulk transfer method developed by Interface Corporation. The Bulk transfer method results in reduced loads for computers and D/A converters, ushering in the next dimension in playback sound quality.
* Dedicated drivers must be installed on your computer when using Bulk Pet.