Singxer SU-6
Jun 13, 2020 at 5:11 AM Post #316 of 1,010
So setting up su-6 is2 to a dac. its out of phase, how do yo set up a su-6 switches. on the bottom. i have a audiobyte black dragon. this is the is2 pin out they sent me.

HDMI_LVDS.png
 
Jun 13, 2020 at 5:26 AM Post #318 of 1,010
I use I2S output and settings: 1-8 OFF, and 9-10 ON
 
Jun 14, 2020 at 8:23 PM Post #320 of 1,010
Well guys I've decided to upgrade from my SU-1 to a Gustad U16 + C16 external reclocker. I got a good deal on a combo unit for around the same price as an SU-6 and the seller said his works without the glitches others have reported with theirs. Some even say that a (properly working) U16 sounds better than the SU-6.
 
Jun 15, 2020 at 2:22 AM Post #321 of 1,010
Well guys I've decided to upgrade from my SU-1 to a Gustad U16 + C16 external reclocker. I got a good deal on a combo unit for around the same price as an SU-6 and the seller said his works without the glitches others have reported with theirs. Some even say that a (properly working) U16 sounds better than the SU-6.
Aww.
 
Jun 15, 2020 at 11:02 AM Post #323 of 1,010
Have you considered the Denafrips Gaia? Personally I’m very happy with the SU-6.
I've run into it and it looks interesting, but I don't think there's anything special about it? Compared to the singxers and other offers.
Could be interesting but unless super positive feedback trickles in, looks like just another DDC to me, personally.
 
Jul 5, 2020 at 5:09 AM Post #324 of 1,010
Anyone here moved from an su-6 to something like the Gaia or di-20he?
 
Jul 26, 2020 at 6:05 AM Post #326 of 1,010
Singxer SU-6 has a design BUG

After exchanging the oscillator of my 004 and two days of burn in, I won'ted to do some listening tests.
So I sat down in my listeningchair everything was silent. I started the player, but no sound :frowning2: . Ohhh, I forgot to plug in the HDMI cable comming from the singxer. I plugged it into the 004, ....... and there it was, a soft hum.
Now you have to know, that I know that very soft hum. It started when I connected chassis GND to PCB GND. But I never could understand where the ground loop was.
Well, until now. I unplugged the HDMI again, and the hum went away. Pluged it back in, an the hum was back.
Pulled out the USB from the singxer, hum was gon. There had to be some kined of connection between the USB input of the singxer and the HDMI output. So I measured it. 0 Ohm between USB socket housing and HDMI socket housing.
I opened up my sand filled singxer, and took a closer look. Indeed, as everybody can see on pictures, the USB section is clearly isolated from the rest of the PCB. BUT ONLY THE MIDDLE 4 PINS OF THE USB PLUG!!!! THE USB PLUG HOUSING IS TIED TO CHASSIS GND!!!!!!!

Almost every USB cable has shielding connected on both sides of the plug, and that's why I had a ground loop.
The solution is to eather isolat the metal part of the USB plug on the singxer side with kapton tape, or open up the connection on the PCB as I decided to do.
That solved a lot of sonic problems. I can bite me in my ..... that I didn't question the SU-6 construction earlier.
 
Jul 27, 2020 at 11:09 PM Post #329 of 1,010

On my SU-6 there is a continuous connection between the metal shroud of the USB connector and the metal shield on the HDMI I2S connector as @Xoverman reported

The SPDIF outputs were still isolated

Wrapping the USB connector with Kapton tape is a better alternative to preserve the unit's warranty in my opinion.

It made quite a difference on my LKS004-->SchittMjolnirV1-->HD800
Nearly all of the vocal sibilance is gone.

The main reason why I got SU-6 was to provide isolation and relocking so this discovery is disappointing.
I checked my SU-1 and it has the same issue.

I2S has a ground connection to the computer which means it is up to the DAC to prevent noise from getting in.

On the LKS004, the metal shell of the HDMI connector is not connected to the chassis ground, the leakage is somewhat contained.
I had previously wrapped the HDMI metal shell on the LKS in Kapton to insulate it from the chassis metalwork, so this also helped but was not enough as the latest experiment shows.

DSCN1292.JPG


DSCN1293.JPG
 
Last edited:

Users who are viewing this thread

Back
Top