Audio-GD DI-20

[tumpux]

If you can solder, you can build one for not much money. I built one myself for about $30 worth of parts. Just need to be a bit careful as the plastics on the plugs can melt easily.

Can we see what it looks like?

 

[Jackula]

Can we see what it looks like?

 

[Oepsie]

If you make your own usb, try without the two power lines.. the di-20 can handle it, and it gives a nice improvement. Much more natural sounding.

 

[ProLoL]

If you make your own usb, try without the two power lines.. the di-20 can handle it, and it gives a nice improvement. Much more natural sounding.

+1

 

[Jackula]

If you make your own usb, try without the two power lines.. the di-20 can handle it, and it gives a nice improvement. Much more natural sounding.

I think you are referring to the power and ground lines, both can be noisy sure, however the 5v is the most noisy and is required for the initial negotiation. Once it's all hooked up you can disconnect the 5v line, but will need to hook it back up whenever you're restarting your streamer or DI.

As for the ground line, as long as it's isolated at the upstream, it won't affect anything because there won't be any current flowing through it.

 

[smodtactical]

Im just using an amazon 8k shielded HDMI cable. I got a chroma 6 coming, looking forward to compare.

 

[DecentLevi]

I've been encountering some chronic issues with 4.07 firmware on my DI-20 98/90M version. Ever since the update, I've had to deal with all kinds of noisefloor and even painfully loud squeeling sounds from the DI-20. I've mentioned it before, hoping it was just on occasional glitch; at the time I posted that it's just a white-noise static background sound occuring sometimes which can be abated by powering down your whole chain and waiting a few minutes; an update on troubleshooting now is that basically changing any front panel settings can resolve it (switch between P - S modes or switching from USB to coax input and back again with a simple push of the "selector" button).

Unfortunately, not only does it sometimes takes multiple times mucking with cycling between front panel settings and returning again, but it happens in a variety of situations:

• initial power-up: often there is a substantial noisefloor akin to white noise behind the music (around 10db of it), only while a 'song' is playing but not when paused so sometimes it takes me a few minutes to even notice it, depending on the dynamic range of the recording I'm listening to; becoming audible during brief gaps between sounds
• DSD (.dsf) playback: this is absolutely NOTORIOUS!! closing the Foobar audio player during DSD playback results in a squeeling sound that can be ear-piercing depending on the volume you're listening to
  - skipping around within a DSD file can cause this too
  - a solution seems to be actually pausing or stopping the playback vs. just closing the media player
• Any random time you resume listening after having the DI-20 already on, there's a chance you will get either full-volume overbearing white noise, or varying amounts like 10db interjected or even perhaps half that and can take a while to notice it's there until you get a momentary break in sounds from your song.
• During a whole session it's liable to go completely unnoticed until there's a completely silent portion on the track you're listening to and with isolating closed-back headphones in a quiet room. This is very hard to detect because it's a very faint background hiss and is one the most frustrating, that you don't even notice this gremlin in the background unless in special circumstances.

This seems to occur approximately 20% of the time when starting up music, about 90% of the time with DSD files (per above detail), and maybe 30% of the time on initial start-up. The main trigger seems to be powering on, switching settings or pressing "stop" on any audio file, though it does seem to behave better when simply pressing pause. Luckily that whole sessions background noise I have also tested to be solve-able with the same setting mucking about as mentioned above, and it takes a special listening environment to be sure it's been mitigated.

Furthermore I've tested the above noise issues to be present on both my Qutest DAC connected via RCA-to-BNC coax cable, as well as with my iFi Micro iDSD Black Label DAC connected by standard RCA coax cable. This issue may be present only with the coax (S/PDIF) connection type, however for me I won't know otherwise because that is the only compatible input type on both of my DACs.

At this point I'm still just dealing with it because in fact, the sound I'm getting from v4.07 is aplomb and quite impressive with every genre, headphone and amp I've paired it with... but come on, this flaw is TOTALLY UNACCEPTABLE and even hazardous to your hearing. I will send Kingwa a link to this post and keep my fingers crossed he can release a firmware update that keeps the same EXTRAORDINARY sound, but with the noisefloor issues solved.

 

[Oepsie]

I think you are referring to the power and ground lines, both can be noisy sure, however the 5v is the most noisy and is required for the initial negotiation. Once it's all hooked up you can disconnect the 5v line, but will need to hook it back up whenever you're restarting your streamer or DI.

As for the ground line, as long as it's isolated at the upstream, it won't affect anything because there won't be any current flowing through it.

I don’t think this is true. I recently bought an sbooster vbus2 which disconnects the 5v as I understand it. Don’t think there is more than one power signal in the usb signal? And it works fine.. No trouble with initial negotiation..

 

[rsbrsvp]

My USB cable is pure silver DUAL HEAD so I can disconnect the v-bus line after I restart my mac mini and just use the data line of the USB.

The V-bus is needed for the initial handshake.

 

[ProLoL]

Im just using an amazon 8k shielded HDMI cable. I got a chroma 6 coming, looking forward to compare.

Got a Chroma 6 0.3m on the way too, would compare.

 

[Jackula]

I don’t think this is true. I recently bought an sbooster vbus2 which disconnects the 5v as I understand it. Don’t think there is more than one power signal in the usb signal? And it works fine.. No trouble with initial negotiation..

The SBooster must allow the 5v to pass through at the start then. As both @rsbrsvp and I indicated, it is required for the handshake.

I have a USPCB which also lets me disconnect the 5v, mine is manual as it will only allow me to disconnect after the handshake.

 

[smodtactical]

Got a Chroma 6 0.3m on the way too, would compare.

I got the 0.5 meter chroma 6 coming.

 

[Oepsie]

The SBooster must allow the 5v to pass through at the start then. As both @rsbrsvp and I indicated, it is required for the handshake.

I have a USPCB which also lets me disconnect the 5v, mine is manual as it will only allow me to disconnect after the handshake.

Ok, sorry for the confusion then. Thought that the vbus actually disconnected the 5v (and that this was the only thing it was doing) - my mistake

But from the sbooster site: https://www.sbooster.com/sbooster-tweaks/vbus2-isolator/

“The Vbus isolator disables the polluted power from the host PC/streamer, therefor it only works with USB audio devices, that do not need power from the host PC or streamer to operate.”

Looking at the usb specification: http://www.interfacebus.com/USB-Interface-Description.html, the 5volt (red) is referred to as the vbus (hence sbooster’s naming of the vbus2). I don’t think the little gadget is capable of switching off the signal after the handshake??

Just trying to understand what is does and how it is different from cutting the 5volt? - I’m a bit confused

 

[Jackula]

Ok, sorry for the confusion then. Thought that the vbus actually disconnected the 5v (and that this was the only thing it was doing) - my mistake

But from the sbooster site: https://www.sbooster.com/sbooster-tweaks/vbus2-isolator/

“The Vbus isolator disables the polluted power from the host PC/streamer, therefor it only works with USB audio devices, that do not need power from the host PC or streamer to operate.”

Looking at the usb specification: http://www.interfacebus.com/USB-Interface-Description.html, the 5volt (red) is referred to as the vbus (hence sbooster’s naming of the vbus2). I don’t think the little gadget is capable of switching off the signal after the handshake??

Just trying to understand what is does and how it is different from cutting the 5volt? - I’m a bit confused

You're not wrong!

Maybe the SBooster doesn't work the way it's described, or your DI is a special unit that doesn't need the 5V

 

[smodtactical]

For those of you doing streamer + di20he. Do you use a cat6 ethernet cable that is unshielded or cat7 Also do you have any special network switch ?