• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is there a difference between a usb 2.0 port and audio grade usb 2.0 port

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Another related question. From a measurement perspective is an AES connection to the DAC similar, better or worse than a connection by USB from the music server?
AES should give you ground loop immunity which would be a plus.
 

tourtrophy

Member
Joined
Apr 17, 2019
Messages
9
Likes
21
The benefit of AES is differential with common mode noise rejection. That 's all. The DAC is still susceptible to the recovered clock jitter. On the other hand, USB can be asynchronous with a local low jitter clock driving the DAC. If the local buffer is sufficient enough, it can communicate to the upstream host using a flow control mechanism to prevent buffer overflow. If implement properly, the DAC processor only needs to use the USB 5V for cable detection to initiate enumeration with the upstream host. In this case, the DAC processor is a self-powered USB device. The 5V detection can be isolated with an opto-coupler chip. If you are concern about the ground noise, you can always use an external galvanic isolator such as the one from iFi Audio (USB certified). So it depends on the circuit design and how the designer chooses the circuit components.
 

L5730

Addicted to Fun and Learning
Joined
Oct 6, 2018
Messages
667
Likes
434
Location
East of England
When I saw this on Gigabyte boards back when the i7 4770k was current, I thought it was quite cool. However, I think it's largely unnecessary.
I suppose USB spec allows for minor modulation of voltage (it should be a constant and flat +5v DC) and other oddness, I mean a keyboard or mouse will probably work without noticeable errors.

If the motherboard is designed properly and a decent PSU is used, then all USB ports should be adequately clean. It then puts more onto the designer of the USB DAC to clean up the power coming in - which they should do anyway as they have no clue what the computer manufacturer is actually doing. So that would all become moot.

No isolated ground, so probably makes little difference, unless the DAC is really susceptible to voltage change.

I'll put the disclaimer that I am, by no measure, an electronics guru.
I would rather use galvanic isolation for USB or keep the host and device tethered with a common ground. From all the stuff I was reading there is a finite window in which the data lines will work, and that is in reference to the ground voltage - which should be the same between host and device. Having too high (or low/negative) voltage on the data lines, as seen by either device or host could cause irreparable damage.
I discovered that the D10 USB DAC will work without the common ground, after enumeration, but it's not really the proper long-term solution.

Galvanic isolation is different.
Also, using I2S from the DAC chip and galvanicly isolating that seems like a good solution in theory, but pricier.
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
The things to worry about.
1, the cleanness of the usb themselves. From psu and local bypass, I don't know if any computer provides clean usb power but the device itself should take care some of that.
2, the interference with each other. I have a few usb audio devices. When I connect any two the same time, there is noise at high frequencies 8khz 16khz. (Audible) At least show up in the measurements with different devices.

I just don't like usb audio that much even though I was the type of guy who think usb is more than enough for audio (no not good enough)

I end up using optical for everything. Or it's possible to use some adapters that can do galvanic isolation and juice up the signal to clean signal. I don't want to recommend this brand but ifi does make these stuff. Not necessarily cheap or expensive. But that's science behind it unlike some snake oil bullshit.
 

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
4,636
Likes
2,809
When I connect any two the same time, there is noise at high frequencies 8khz 16khz. (Audible) At least show up in the measurements with different devices.

I'm always looking out for 8kHz USB packet noise in @amirm 's measurements.

I know this issue needs to largely be tackled in the DAC USB input design but Archimago found a Corning Optical USB cable helped in his system. I got one and have enjoyed it. I mainly use it because the 10m length is useful in my setup...

"Compared to the spectrum above, although the technique did not completely remove the 8kHz "packet noise", it is now down to about -120dB, plus the overall noise floor is down another 5-10dB. Basically, there has been a >20dB suppression of the packet noise which has now rendered it inaudible at normal listening levels. Also, no more audible "processing noise" when I open or close programs like JRiver, or do other tasks on the computer."

http://archimago.blogspot.com/2015/05/measurements-corning-usb-3-optical.html

The power and ground wire is very very thin and the shortest Corning cable is quite long, at 10m. Data is optically transmitted.
 
Last edited:
  • Like
Reactions: DDF

tourtrophy

Member
Joined
Apr 17, 2019
Messages
9
Likes
21
If the products carry USB logo, chances are it should pass the signal integrity tests for the data lines. In USB2 electrical compliance testing. they check the eye diagrams and also measure the rise time/fall time with a short channel model on a scope (meaning short cable). If you have a 3-meter or shorter cable, you might see more ringing at the far end of the channel/cable. I have seen products that fail on short channel but not long channel before. This is undesirable for EMI, crosstalk, and such. VBUS is traveling along the same lengthy cable in parallel. VBUS itself from the USB host is also likely to be noisy as it is typically shared with some other digital circuits on the motherboard and probably sourced from a switch mode regulator. The best design is to have the DAC to be self-powered. The XMOS chip only needs to detect cable insertion for VBUS presence. Self-powered VBUS is a little bit more costly since you need to add power regulator(s) and provide a clean source. Once you add galvanic isolation for the ground, you are pretty much isolated from the noisy host motherboard.
 
  • Like
Reactions: DDF
Top Bottom