DIY ES9038pro Board

[erikovsky]

Hi all,

I did a comparison chart, showing phase noise of NZ2520SD, CCHD-957, and the new DuCULoN from NDK.

NZ2520SD"A" is a better version of NZ2520SD, reaches -172dBc/Hz at 100KHz while cost ~ $10 USD.

So I think NZ2520SD"A has better cost-effectiveness than CCHD-957 (3 times cost).

 

[erikovsky]

To correct the chart: NZ2520SDA
This is the right version, I put NZ2520SDA(as NZ2016) into comparison.
Therefore, as we can see, NZ2520SDA has lower phase noise figure than CCHD-957 all the way to 1MHz, as low as -172 dBc/Hz.

 

[Markw4]

To correct the chart: NZ2520SDA
This is the right version, I put NZ2520SDA(as NZ2016) into comparison.
Therefore, as we can see, NZ2520SDA has lower phase noise figure than CCHD-957 all the way to 1MHz, as low as -172 dBc/Hz.

NDA makes no guarantees in their specs that such performance can be relied upon. They won't even publish an example phase plot in an English language document.

 

[wealas]

Some independent measurements here: analog research NDK low jitter clocks for digital audio ? Not the SDA version but better than nothing.

 

[synonymous]

Most people who try everything the chips can do find synchronously clocked DSD over galvanically isolated USB, and with reclocking following isolation stage to sound best.

It is also my intention to not need any wires (interconnects) between components. The RPI compute module can be used for it's USB pins without need for an interconnect.

 

[erikovsky]

NDA makes no guarantees in their specs that such performance can be relied upon. They won't even publish an example phase plot in an English language document.

According to analog research NDK low jitter clocks for digital audio, even 2520SD non-A version is
"15 dB better than Crystek", not to mention 2520SD"A".

Plus, the 2520SDA costs 1/3 of CCHD-957, I don't see any reason to use CCHD-957, just because it's famous or?

And, yes, for sure CCHD-957 gives some promising specs, like <3ppm in the first year. But 3 times of the cost?! NO!

 

[androa76]

Any source of 100mhz NDK clock? I see till 50mhz max range. For usb boards only .

 

[erikovsky]

Any source of 100mhz NDK clock? I see till 50mhz max range. For usb boards only .

You can get NZ2520SDA with fair prices from DIYINHK, but sadly, frequencies like 24MHZ, 50MHZ, 100MHZ are not available.

Especially 100MHz, which is needed in ES9038 design. Crystek CCHD has 100MHz, that the only reason I would go for Crystek

 

[Markw4]

Any source of 100mhz NDK clock? I see till 50mhz max range. For usb boards only .

NDK makes an NZ2520SD 80MHz clock (no "A" suffix).

Regarding the A-suffix parts, they appear to be selected or binned from non-A production parts.

The very best NZ2520SD parts people ever found were around 15dB better than what Crystek claims for 957 parts, but there were also typically some NZ2520SD duds on every reel that were junk. Therefore, each one had to be tested by the user, or by some 3rd party lab, or people could just take their chances. Now that A-suffix parts are binned out, at least you don't get the worst ones (presumably), but the lowest offset frequency NDK is willing to spec on the data sheet is still 1kHz offset. To me, that means you can still get one with higher than average 1/f corner frequency. Of course, not specifying performance means the clocks can sell for less, since it would cost time and money for NDK to fully test each one.