Gustard U18 DDC Measurements

Content and analysis made possible thanks to support from https://headphones.com, and https://patreon.com/goldensound supporters

This unit was loaned to me by Aoshida Audio for testing and review

UPDATE:
This post previously said that there was no capability to adjust I2S pinout. This was incorrect. I had mistakenly understood that the I2S pin adjustment only changed the DSD flag pin.
There are in fact 4 different I2S pinouts selectable on the U18.

ChitChat:

The Gustard U18 is a DDC retailing at $499 and intended to be paired with the X18 DAC, though of course you can pair it with anything you like thanks to its multitude of output options.
This is one of the few DDCs that supports 768khz output over I2S, most are limited to 192khz or 384khz, so it could be an ideal option for those looking to use HQPlayer or PGGB through a DDC.

There are no SPDIF inputs, only USB. So this cannot be used as an SPDIF reclocker.

The U18 features optical and coaxial SPDIF output, AES, and an LVDS (HDMI) I2S output using the gustard pinout standard.
There is also the capability to change between 4 different I2S pinouts, so the U18 should work with most DACs but double check your DAC pinout just in case, as some have unorthodox layouts that may not match any of these.

There is also a 10Mhz clock input, and the Gustard ‘K2’ clock synthesizer, allowing you to use a 10Mhz external reference clock with the U18.

Internally there is a linear power supply feeding the main unit itself.

We can see the galvanic isolation barrier in yellow on the left hand side with isolator ICs bridging across to the mainboard, ensuring that no noise can be passed from the USB input to the mainboard and outputs themselves. Also to note, the USB port has a rubber shield to prevent it contacting the chassis and therefore connecting the grounds, breaking galvanic isolation.
This is something that surprisingly quite a few manufacturers don’t seem to do!

On the other side of the isolation barrier are components for filtering and regulation of DC power, and the two Accusilicon AS338 clocks which feed the main FPGA.

Overall a very nicely laid out device with excellent external build quality too.
Lets have a look at how it performs.

Measurements:

Test Setup:

– Audio Precision APx555 B-Series Analyzer
– AudioQuest Carbon Coaxial Cable and AES cable(1.5m)
– Intel PC as USB source
– 44.1khz and 48khz real music played through device during measurement
– Coax output used unless otherwise specified
– Device was ‘burned in’ for 100 hours before testing

Jitter:

To measure jitter, the device is connected to the digital input of the APx555, and the analyzer is set to analyze jitter, not audio content.
These measurements do not show audio/analog info, but instead show the spectrum of jitter, ie: time-domain inconsistencies.
This is the primary factor that a good streamer or DDC will seek to improve.

Jitter performance is not particularly impressive unfortunately. And it is beaten comfortably by other DDCs like the KTE SU2, SU6, DI20HE, and in fact the Pi2AES has lower jitter in every regard too.

Jitter is slightly poorer with 44.1khz which is typical on USB devices due to the clock rate not being perfectly divisible into the 48Mhz clock rate of USB.

Given this performance I hooked the U18 up over I2S to the X18 and compared the jitter performance vs using its own internal USB.

When using the X18’s internal USB there are some very low level jitter spurs at about 1khz, 1.5khz and 3khz either side of the fundamental. When using the U18, the 1500hz and 3khz spurs are gone, but the 1khz spurs remain, and larger low frequency sidebands are added.

It’s tough to say whether this is overall better or worse performance as it’s mostly just changed the structure of jitter instead of outright reducing it. But it’s not really an outright improvement.

These sidebands are actually present regardless of the I2S source I used, so it seems this may be a limitation of the X18’s own I2S input.
However using a lower jitter DDC such as the KTE SU2 did result in lower level sidebands than the U18. So even with this limitation, using a lower jitter DDC still gets you better results.

Noise levels on the U18 are again ok but not great, and is beaten by most other DDCs I’ve tested if going by RMS/Peak values.

Overall, the U18 is a good DDC and will provide full galvanic isolation, but there are other DDCs for similar money that perform better.
I would personally only recommend going for the U18 if you’re happy to sacrifice a little bit of performance for the aesthetic of a matching stack (which is a completely valid reason IMO, there’s more to the enjoyment of your setup than numbers.)