Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: high-profile ABX test of ethernet cables (Read 14784 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

high-profile ABX test of ethernet cables

Reply #25
Just to report things i did read lately about this stuff.
Seems like the golden ear brigade agrees on that the ethernet cable is always clearly disturbing the sound of any audio gear nearby. Of course no one knows yet what to measure but it must be some packet noise or alike more than -160dB down.
I even learned that the 8kHz packet noise Archimago measured at his blog with USB proves exactly this because ethernet uses also packets!
If you want to learn from the thread i have this revolutionary knowledge from just look at the usual place


I have a number of USB DACs where SNR measurements detect 8 KHz packet noise.

Interestingly enough some of them can be "cured" by optimizing the connection from the analog output of the DAC to the rest of the system.


Those don't sound like very good DAC's.  What kind of noise floor are we talking about?


Well over 100 dB below FS.

Quote
Do you mean a cable "upgrade?"


The DACs I'm talking about have RCA jack unbalanced outputs. My test equipment has TRS or XLR balanced inputs.  There are a number of different ways to make an unbalanced to balanced adapter, and some work better than others in terms of rejecting the usual hash that seems to surround USB converters.


So, the DAC is passing the USB noise without affecting the DAC's performance, and then the balanced connection rejects the noise?  Would ferrites on the USB cable help?

high-profile ABX test of ethernet cables

Reply #26



Do you mean a cable "upgrade?"


The DACs I'm talking about have RCA jack unbalanced outputs. My test equipment has TRS or XLR balanced inputs.  There are a number of different ways to make an unbalanced to balanced adapter, and some work better than others in terms of rejecting the usual hash that seems to surround USB converters.


So, the DAC is passing the USB noise without affecting the DAC's performance, and then the balanced connection rejects the noise?  Would ferrites on the USB cable help?


With the right hook up of the balanced input to the unbalanced output, the USB noise is rejected.

The frequency of the noise is 8 KHz, which ferrite beads generally don't do much about.  Too low of a frequency.

high-profile ABX test of ethernet cables

Reply #27
The frequency of the noise is 8 KHz, which ferrite beads generally don't do much about.  Too low of a frequency.

8 kHz may just be the amplitude modulation of a much higher frequency, so ferrite might actually work.


high-profile ABX test of ethernet cables

Reply #29
Please note that USB high speed sends in 125us intervals (= 8 kHz), that different "normal" (copper) USB cables don't change anything so fancy normal ethernet cables won't either, that Archimago was not measuring the output of the DAC and that he was using balanced interconnects ...
"I hear it when I see it."


high-profile ABX test of ethernet cables

Reply #31
8kHz USB noise investigation by Archimago


Good reference that I was aware of when I wrote my last post.

Relevant quote from the above:

"For example, it was serendipitous that the Emotiva XSP-1 pre-amp had such a noise-sensitive unbalanced "HT Bypass" input that I was able to detect noise and measure it previously."

 

high-profile ABX test of ethernet cables

Reply #32
I have a number of USB DACs where SNR measurements detect 8 KHz packet noise.


Can you show us examples?
"I hear it when I see it."


high-profile ABX test of ethernet cables

Reply #34
Thanks. I've never heard of this China DAC. In other words noise makes it from the digital part into the unbalanced analog output/ground. Looks like improper power supplies/shielding/grounding - but it's good that this common mode noise can be rejected with the right cable into a balanced input.

I think linking to Rane doesn't hurt:
Rane - Sound System Interconnection


So I guess that Emotiva pre doesn't properly reject common mode noise in "HT bypass" mode. They probably messed up the grounding.
"I hear it when I see it."

high-profile ABX test of ethernet cables

Reply #35
Thanks. I've never heard of this China DAC.


That's the nature of eBay "Specials".

The DAC  chip they use can be configured a number of different ways by means of programming pins, but as sold its performance is like this:

Uploads forum link to measured FR.

Quote
In other words noise makes it from the digital part into the unbalanced analog output/ground. Looks like improper power supplies/shielding/grounding - but it's good that this common mode noise can be rejected with the right cable into a balanced input.


Agreed.

Quote
I think linking to Rane doesn't hurt:
Rane - Sound System Interconnection


Yes, the drawing I showed was a modificatation of circuit 18 from there.

Quote
So I guess that Emotiva pre doesn't properly reject common mode noise in "HT bypass" mode. They probably messed up the grounding.


The "HT Bypass" input has no common mode rejection because it is unbalanced.

Quoting Archimago again:

"For example, it was serendipitous that the Emotiva XSP-1 pre-amp had such a noise-sensitive unbalanced "HT Bypass" input that I was able to detect noise and measure it previously."

high-profile ABX test of ethernet cables

Reply #36
As you can see from his blog post, his configuration with noise looked like this: DAC -> balanced cable -> pre (bypass mode) -> balanced cable -> ADC.

But look at the manual:
Quote
2. Home Theater Bypass Button
When you press this button, the source connected to the Home Theater Inputs is routed directly
to the Main Outputs on the XSP-1, bypassing all controls including the level control and bass
management. The display will indicate Home Theater Bypass when this is active.(Conversion
between balanced and unbalanced is still performed, and the XSP-1 must be On in order for the
outputs to be active.)

Quote
16. Home Theater Inputs
Balanced and unbalanced inputs. Connect either balanced or unbalanced line level outputs
from your home theater preamp / processor or receiver to these inputs. DO NOT connect the
speaker level outputs from a home theater receiver to these inputs. When the HT Bypass
Input is selected, signal from the Home Theater Left and Right Inputs is routed directly to the
Main Outputs, and signal from the Home Theater Sub Input is routed directly to the Subwoofer
Outputs.
Note: In HT Bypass mode, ALL controls on the XSP-1 are bypassed, including the Volume control
and all bass management functions.
Note: There is only one set of Home Theater Inputs; both balanced and unbalanced input
connections are offered for flexibility.


So maybe they just cheaped out on those balanced HT inputs.
"I hear it when I see it."