Thursday, July 7, 2011

Aries Cerat Kassandra Dac

In the audio industry today, the market is in abandon of digital processors, with a new dac presented each week, in a race of finding the best converter. A race of numbers, with THD of -90db, oversampling @ radio frequencies and about 140db dynamic range. Yet, many audiophiles change dacs as soon as the next one arrives, without ever finding the one sounding “just” right. In the process of development of our reference system, we were also searching for the best available converter. Yet, no converter delivered the sound we expected, as a reference converter. Every dac presented an artificial feel in every recording, and almost all presented compressed dynamics and poor flow of music, easily noticed with un-compressed recordings and ultra high sensitivity horn systems. So, we began designing the best possible digital processor.

Delta sigma dacs are todays’ standard for digital processors. Not because they present better audio quality, but because of availability and ease of implementation of digital filters. Filters that became more and more complex, only because of the artifacts of delta sigma modulation. R2R converters are only available in instrumentation converters because of their higher price. They are used for their supreme settling time and absence of glitch artifacts. We chose to use the AD1865, a 18bit R2R converter, probably the best sounding audio dac presented up to date and especially in the version we use, the AD1865N-K. We went a bit further and used 24 of them, 12 converters per channel. Paralleling converters improve the total SNR improves linearity and channel separation.

Dynamics and micro-details improve so much, so you can easily say that it is  another converter altogether. The 12 converters in each channel work in complimentary mode, so that 6 converters are inverted.
The 12 converters working in current mode use an I/V transformer so that the current differential is converted in single ended voltage output. The analog stage is implemented by a single super-tube, the E280F, working in triode mode, loaded with a step down output transformer. The tube’s grid is driven by the secondary of the I/V transformer via an adjustable bias supply. The tube power supply is an oversized choke regulated supply, making sure that the tube is up to the task of following the converter’s flashing dynamics. Furthermore, the stepdown transformer reduces the output impedance to 60ohms, with a maximum swing @ full scale of 30Vpp sine wave.

The converter is equipped with an internal super-clock, which reclocks the converters, making the processor virtually immune to transport changes. The jitter is attenuated to < 80psrms , whatever the jitter level is present at the transformer coupled spdif input. With the internal clock, there is no need for a work clock input. All digital logic is designed and optimized in-house. The super-clock reclocker can be bypassed on the fly by remote control, so you can hear the effect of jitter elimination in your system. The converter’s power supply is a complex system of a total of 8 torroidal transformers, for full separation of digital section, analog supplies, clock supply and the tube circuits. A total of  > 2F of capacitance is present in the system, with more than adequate local de-coupling, using top line capacitor, such as OSCONs, WIMAs and Nichicons.