My next upgrade

@baylinor  

10Mhz clocks were never intended to improve jitter performance. They were intended to keep many devices in sync and prevent ‘clock drift’ over time in professional situations ...

https://goldensound.audio/2022/04/14/gustard-x26-pro-measurements-f20-firmware-10mhz-clock/

In OP's situation, the proper way to use an external clock, as suggested above that has been used by several users including @sls883, is to employ an external streamer up front, connect it to a DDC via USB, and then feed the signal to the DAC via I²S. A 10 MHz clock should be provided to both the DDC and the DAC. One could potentially eliminate the DDC if the external streamer or digital transport supports both I²S output and a 10 MHz clock input. Whether a sine wave or square wave works better is another subject.

@lanx0003 

Actually, your info is not accurate. If the master clock is connected to the DDC, the I2s connection transfers that clock signal automatically to the dac. Hence the master clock connection to the dac is unnecessary. My setup has the master clock connected to the ethernet switch and the DDC. Much smarter way of using your clock.

@baylinor  First of all, we're discussing the OP's situation, and it seems you're suggesting feeding an external clock to the DAC only. I just provided a technical source explaining that using a 10MHz clock alone may not improve performance. The OP doesn't have an external streamer, so a DDC cannot be utilized.

Second, a more contentious topic: if you have a network switch, by no means should you connect it to an external clock. The Gustard C16 or C18 provides only two sine wave and two square wave outputs, limiting its usefulness beyond two devices—though other clock generators might offer more outputs. Still, synchronizing multiple devices is the intended purpose of such a clock.

Now, the more debated issue is whether the DAC itself needs to be clocked. Many people seem to assume that the DAC automatically slaves to the DDC’s clock when connected via I²S, but that’s not necessarily true. What I can say is this: it depends—not all DACs behave the same way. It depends on the DAC’s internal design. Some DACs are designed to slave to the incoming clock signal via I²S, while others use a PLL (Phase-Locked Loop) mechanism to regenerate a cleaner clock from the incoming signal for ASRC (Asynchronous Sample Rate Conversion).

The Harmony (micro) DAC then has a unique design that allows the user to select between using the I²S clock or the DAC's internal clock. This is advantageous if the DAC features a higher-grade internal clock, such as the OCXO used in the Gustard X30 or Denafrips Terminator 15th. In such cases, feeding an external clock for synchronization is considered beneficial. Some user here has already tried this and reported positive results.

Look at the description of his equipment up top. He has a DDC.

Anyway, this blabbing is not for me. My attempt to help was strictly about square vs sine wave. And you decide to write a book about a completely  different subject. Have a good day.

You said

Hence the master clock connection to the dac is unnecessary.

I tried to explain why it might be necessary—clear, concise, and completely on-topic. Too bad you didn't understand it.