Cayin announced the T01 Audio Motherboard on 8 August, We have two public reviews from Twister6 and Headfonics already, and a number of N6ii users in this forum has shared their impression, so I assume we know quite a bit about the T01 already, As usual, we leave the subjective listening impression to professional reviewers and users, we shall focus on the technical and circuit design of T01 only.
Burr Brown PCM1792A DAC
Cayin is one of the most experienced high-end audio manufacturer using PCM1792/1792A DAC family. As soon as Burr Brown made the original PCM1792 engineering sample available, we studied and implemented in into our flagship CDT-17A CD player. The revised PCM1792A was announced back in 2009 and Cayin used that is our GenII flagship CD player CDT-17A MK2 as well. When we decided to venture into portable audio and develop our first DAP around 2013/2014, we have considered other DAC options such as ESS9018 but at the end of the day we decided to stick with PCM1792A because the sound are closer to our preferred audio signature, and N6 remains the only DAP using DUAL PCM1792A till today.
The T01 Audio Motherboard adopted the same DAC architecture with N6 using two PCM1792A in Mono mode, however the DAC chipsets in T01 is operated in DSD mode while the PCM1792A in N6ii were operated in PCM mode. We would like to highlight the following details regarding the digital decoding capability of T01:
- The FPGA of N6ii will feed the unaltered PCM (upto 32Bit/384kHz) and DSD (upto 1Bit/11.2896MHz) bitstream to T01 Audio Motherboard. We didn't perform any upsampling or downsampling at FPGA stage.
- When you playback a Hi-Res PCM file up to 24Bit/192kHz, T01 will decode the music file directly and display the sampling frequency on status bar.
- When you playback a Hi-Res PCM file up to 32Bit/384kHz, T01 will decode the music file directly and display 192kHz sampling frequency on status bar suggesting T01 has downsampled the digital signal before decoding
- When you playback a DSD64, DSD128 DSD256 music file, T01 will decode the music file directly and display 2.8224Mhz, 5.6448MHz and 11,2896MHz on status bar accordingly, suggesting T01 has decoded the digital signal directly (or described as hardware decoded).
The following tables (as extracted from the
datasheet) described the digital capability of PCM1792A in detail:
The DSD mode of PCM1792A operates in 2.8224Mhz and supports system clock frequency from 2.8224MHz to 11.2896MHz, which is equivalent to DSD64 to DSD256. Burr Brown/Texas Instrument didn't disclose the DSD decoding algorithm so there is no way to figure out if there is any resampling were involved within the chipset, it is non-conclusive at this stage because it didn't display a lower sampling frequency from its status.
Current Mode DAC Output
The PCM1792A is a current mode output DAC, it is designed to supply a current into a virtual ground as output and required an I/V converter Op-Amp to convert the current output to voltage output so that it can be feed into subsequent analogue amplification circuit. It is different from the Voltage mode output such as the highly integrated AK4497 which has a build-in op-amp (in simplified terms) to generate a voltage output directly.
Current mode output didn't necessarily sound better then Voltage mode output, but the external IV converter will in most case, outperformed the built-in converters of voltage mode DAC, and offered more room to experienced developer the find tune the sound signature of the audio circuit. For instance, T01 uses FOUR audio grade op-amp to serve as the fully differentiate I/V converter circuit and this is part of the reason why T01 has excellent dynamic and holographic soundstage in its overall presentation.
High Precision Low Noise Digitally controlled Analog Volume Control
The PCM1792A didn't have an excellent built-in digital volume control like modern ESS and AKM DAC chipset, so we need to implement the volume control at analogue level. While there are numerous options such as carbon film potentiometer, Cayin has opted the more expensive (and power consuming) microprocessor approach by using TWO digitally controlled stereo analog volume controller (PGA2311U). These are highly programmable, high precision, low noise, low cross-talk volume control device . We can eliminate channel in-balance even at very low level, and offers 99 steps of volume control down to 0.5dB per step at our discretion. The original N6 also use one unit of PGA2311 but since the T01 is a fully balanced design, we need to use two stereo volume control and that is one of the reason why T01 has a shorter battery duration.
Fully differential Design with Parallel Headphone Amplification
The T01 is a fully differentially design from DAC all the way to headphone amplification. It share similar final headphone output stage with A01 Audio Motherboard which employs 4 pieced of OPA1622 in parallel amplification configuration and offers upto 530mW per channel at 32ohm loading from its 4.4mm balanced headphone output. Please be reminded that rated power output with load is more reliable and meaningful performance index when compare to unloaded output rating. Parallel amplification increases the output current capability and lower the output impedance. In practical, these improvements will translate into better control, ability to sustain a heavier loading, and drive most portable headphones with ease and offers lots of headroom.
These PCB photos are taken from a real-life production T01 Audio Motherboard, itis jam-paced on both side of PCB, our Engineers has explore the physical dimension of N6ii Motherboard to a fairly extreme level. We take a very different approach in our modular concept: each Audio Motherboard is NOT an upgrade from previous model, we offers alternatives to different users, or users with different needs at different time. We think it is unreasonable (and money grabbing) if we offer the T01 as an upgrade to N6ii with A01 within 3 months of product launch. It is NOT, it is a different audio implementation, with different sound signature (and compromises) at the same quality level, and this will be our philosophy when we develop subsequent Audio Motherboards.