[mmerrill99]

There's even a thread here:
http://www.head-fi.org/t/131544/ground-lifting-a-usb-cable

about lifting the ground and things still working, although I'm a little confused about if the shield ground was cut.   Where in any of this is though is the justification for a $500 usb cable that does NOT isolate?

Yes, I've seen this done before & tried it myself but what is happening by cutting the ground wire in a USB cable is that the USB ground is now finding an alternative route back to the USB sending device - possibly through the shield or as appears to be happening in the link, through the mains ground.

So what we have is noise getting through to the output of a USB DAC - is this USB Audiophile DAC therefore broken? Or is it his Dell laptop that's broken? Or is it both?

How would these "armchair designers" suggest he should fix this?

 

[mmerrill99]

A ferrite core can dissipate common mode (only) energy into actual heat.  Non ferrous shielding can't do that.  The only way to fix a ground loop is to remove it or live it with.

Yes 

Removing it means isolating the usb connection.  That's it.  Nothing subtle about it.

What's subtle is knowing you have a ground loop if it's not an obvious hum or whine you are hearing. These obviously audible ground noise issues occur when excessive noise is on the ground. With lesser amounts of noise (or different spectrum of noise), more subtle forms of audible issues arise - flattening of sound stage, deadening of dynamics (I mentioned all these before). Most people don't recognise these issues until they are removed & they hear the sound with the ground noise eliminated. Now before anyone starts on me - these are my experience, not marketing BS regurgitated.

And that's why I said most current USB dACs are USB 2 high-speed devices for which only a small number of isolating devices exist - the Intona or Adnaco Optical or Corning optical. There aren't many choices & these devices cost > $400 AFAIK?  

I'm not thinking a ferrite core will do much, and I'm not keen on other shielding either, but it would be interesting to see in an actual test setup like that one where there is a measurable effect.  And who cares if it "should" be there.  If we're talking about $20 fixes, we're not relegated to discussion of $1000 DACS only.

My initial post on all of this was a direct answer to the question posed in the title "Why USB cables can make a Difference" - I was simply giving the technical reasons why they can make a difference. Is there a better solution? Sure but many don't want to go down the route of isolation or of tracking down ground loops & fixing them in other ways.

If people don't understand the technical explanations given then why argue - ask questions & inform yourself, instead of the knee jerk reactions seen on this thread where good technical info is met with personal abuse.

 

[sonitus mirus]

Yes 
What's subtle is knowing you have a ground loop if it's not an obvious hum or whine you are hearing. These obviously audible ground noise issues occur when excessive noise is on the ground. With lesser amounts of noise (or different spectrum of noise), more subtle forms of audible issues arise - flattening of sound stage, deadening of dynamics (I mentioned all these before). Most people don't recognise these issues until they are removed & they hear the sound with the ground noise eliminated. Now before anyone starts on me - these are my experience, not marketing BS regurgitated.

And that's why I said most current USB dACs are USB 2 high-speed devices for which only a small number of isolating devices exist - the Intona or Adnaco Optical or Corning optical. There aren't many choices & these devices cost > $400 AFAIK?  
My initial post on all of this was a direct answer to the question posed in the title "Why USB cables can make a Difference" - I was simply giving the technical reasons why they can make a difference. Is there a better solution? Sure but many don't want to go down the route of isolation or of tracking down ground loops & fixing them in other ways.

If people don't understand the technical explanations given then why argue - ask questions & inform yourself, instead of the knee jerk reactions seen on this thread where good technical info is met with personal abuse.

 
How is subtle noise from a ground loop flattening sound stage and deadening dynamics?  I can't find a direct correlation to this particular noise causing these effects.  The noise must be measured, surely, and what would make this noise different from any other noise?  This is all from a ground loop?
 
I thought this discussion was about USB cables.  It only confuses things when the replies dig into USB power isolation or exotic hybrid USB/optical connections.  I suppose there is no place else to turn when the USB cables don't change the sound.

 

[mmerrill99]

How is subtle noise from a ground loop flattening sound stage and deadening dynamics?  I can't find a direct correlation to this particular noise causing these effects.

I'm not sure of the exact mechanism but Ben Duncan in his book "High Performance Audio Power Amplifiers" pg 441 states " Modulation Noise: an error or noise signal that is only present concurrent with programme. It may therefore not be audible for what it is although its detail-masking effects may be quiet audible" I would add to this statement that it modulates in correlation to the programme/signal as this also needs to be stated.  

The noise must be measured, surely, and what would make this noise different from any other noise?  This is all from a ground loop?

There are different forms of noise - noise which is unvarying with signal, which is just perceived as a noise floor & can easily be listened through. Think of Vinyl or tape playback - the fixed background noise is evident but easily ignored.
Modulated noise fluctuates with the music signal & would seem to be more noticeable, as a result, not directly but as a detail-masking effect, as Ben Duncan states. To me detail-masking will result in loss of perceived dynamics.

I thought this discussion was about USB cables.  It only confuses things when the replies dig into USB power isolation or exotic hybrid USB/optical connections.  I suppose there is no place else to turn when the USB cables don't change the sound.

If you don't appreciate that USB cables don't just deliver bits then you won't understand any of what is being discussed, I guess?

 

[sonitus mirus]

I'm not sure of the exact mechanism but Ben Duncan in his book "High Performance Audio Power Amplifiers" pg 441 states " Modulation Noise: an error or noise signal that is only present concurrent with programme. It may therefore not be audible for what it is although its detail-masking effects may be quiet audible" I would add to this statement that it modulates in correlation to the programme/signal as this also needs to be stated.  
There are different forms of noise - noise which is unvarying with signal, which is just perceived as a noise floor & can easily be listened through. Think of Vinyl or tape playback - the fixed background noise is evident but easily ignored.
Modulated noise fluctuates with the music signal & would seem to be more noticeable, as a result, not directly but as a detail-masking effect, as Ben Duncan states. To me detail-masking will result in loss of perceived dynamics.
If you don't appreciate that USB cables don't just deliver bits then you won't understand any of what is being discussed, I guess?

 
You were talking about noise from a ground loop.  That is the specific noise I was referencing.  Also, I'm only interested in cables made to USB standards.  There are many USB related items, but I see no reason to step away from certified USB cables until we establish if there is any possible way to show that differences are being heard with the various cables being offered that cost hundreds of dollars or more.

 

[mmerrill99]

I'm not sure of the exact mechanism but Ben Duncan in his book "High Performance Audio Power Amplifiers" pg 441 states " Modulation Noise: an error or noise signal that is only present concurrent with programme. It may therefore not be audible for what it is although its detail-masking effects may be quiet audible" I would add to this statement that it modulates in correlation to the programme/signal as this also needs to be stated.  

There are different forms of noise - noise which is unvarying with signal, which is just perceived as a noise floor

You were talking about noise from a ground loop.  That is the specific noise I was referencing.

We seem to be talking past one another or failing to communicate. A ground loop is a voltage differential between the grounds of two devices. Noise that appears on this ground connection can be many forms - noise of 50/60Hz spectrum from mains rectification, RF noise from various origins, including SMPSes, etc. I'm suggesting, as others do also, that noise which modulates in correlation with the music signal has a detail-masking effect & is perceptually perceived as a reduction in dynamics & flattening of sound stage.

 Also, I'm only interested in cables made to USB standards.  There are many USB related items, but I see no reason to step away from certified USB cables until we establish if there is any possible way to show that differences are being heard with the various cables being offered that cost hundreds of dollars or more.

So you believe that USB cables only deliver bits & don't connect the grounds?
Certified USB cables only show that they do not distort the squarewave digital signal beyond the agreed specifications threshold - it offers no assurances about any possible ground loop issues.

 

[sonitus mirus]

We seem to be talking past one another or failing to communicate. A ground loop is a voltage differential between the grounds of two devices. Noise that appears on this ground connection can be many forms - noise of 50/60Hz spectrum from mains rectification, RF noise from various origins, including SMPSes, etc. I'm suggesting, as others do also, that noise which modulates in correlation with the music signal has a detail-masking effect & is perceptually perceived as a reduction in dynamics & flattening of sound stage.
So you believe that USB cables only deliver bits & don't connect the grounds?
Certified USB cables only show that they do not distort the squarewave digital signal beyond the agreed specifications threshold - it offers no assurances about any possible ground loop issues.

 
I think what most are interested in finding out is how a typical 3' USB cable from Blue Jeans or Monoprice would sound inferior to similar cables as the Nordost Blue Heaven USB cable or an Audioquest Coffee USB cable that cost 100 times more.
 
Personally, I would like to know more about the differences that are mentioned in the sales pitch for the Audioquest Coffee USB cable.  Where is the data to support this notion?  Is it simply customer feedback and pro-audio reviewers' personal opinions?  Why aren't they sure how it is possible to hear differences that are not imagined?  Are we back to magic again?   
 
http://www.musicdirect.com/p-15065-audioquest-coffee-usb-b-plug-cable.aspx

Digital Audio over USB
There are significant, audible differences between USB cables. We're not sure how this is possible, since USB is purported to be a purely digital interface, however, the sonic differences are repeatable and consistent from system to system. A better USB cable makes a better digital audio cable, period.

 

[sonitus mirus]

So you believe that USB cables only deliver bits & don't connect the grounds?
 

 
No, for some of my applications, the USB cable provides both a 5V DC with up to 500mA power and data to my DACs.  I don't hear any noise on the analog outputs of my DACs when using different computers as sources.  I have not taken any direct measurements, though, as there does not appear to be any problems that would warrant further investigation.  I've used many different USB cables from 1-6 feet in length, though none cost more that $20.  

 

[mmerrill99]

Here's what Rob Watts (famous DAC designer) has to say about Noise floor modulation

No power amp or pre-amp "likes" a full bandwidth signal - not if that signal has out of band noise. Out of band noise (RF noise) makes the SQ brighter as it induces noise floor modulation - the noise pumps up and down with the wanted music signal. The brain detects this noise, and can't separate it from the wanted signal, and as noise sounds bright (hiss) it adds an edge to the perceived timbre of the instrument.

The technical reason why its important is down to the fact that audio electronics (whatever flavour) is non-linear at RF frequencies, and random RF noise inter-modulates with the wanted music signal, creating inter-modulation products that is within the audio bandwidth - hence noise that pumps up and down with the wanted signal. I have measured this problem countless times, so its a very real problem.

The brain is extremely sensitive to this issue - and when you remove noise floor modulation, it sounds a lot smoother and darker, with better instrument separation and focus. ............

And this one specifically what he has to say about USB cables & why they might make a difference

At the risk of being flamed, I don't see how the composition of the USB cable wire can add warmth to digital data from whatever device is being used as a transport

I understand those concerns to - after all the data is the same. But there are solid scientific reasons why they can make a difference.

In the 1980's, people started talking about mains cables making a difference to the sound quality - and I didn't believe it either - particularly as my pre-amp had 300 dB of PSU rejection in the power supply. But I did a listening test, and yes I could hear a difference. Frankly I still could not believe the evidence of my own ears, so did a blind listening test with my girl friend. She reported exactly the same observation - mains cables did make a difference to SQ.

To cut a long story short, I proved the problem was down to RF noise. RF noise inter-modulates with the wanted audio signal within the analogue electronics, and if the RF noise is random, then the distortion is random too and you get a increase in noise floor with signal. This increase in noise floor is noise floor modulation, and the brain is very sensitive to it; you can perceive tiny amounts of noise floor modulation as a brightening or hardening of the sound. By tiny I mean the noise floor modulation needs to be well below -200 dB, so the brain is very sensitive to it. With the right test equipment, you (APX5555 is only test equipment that has no innate noise floor modulation) can easily measure the effect.

The RF characteristics of the cable can change the RF noise that gets injected into Mojo's ground plane, and this is the mechanism for changes in smoothness. You may say why can't you make it insensitive to it; well I go to silly lengths to RF filter and decouple, and use dual solid ground planes on the PCB, but you can't remove the problem. For Dave, Hugo TT and 2 Qute I have galvanic isolation, and this eliminates the problem (along with other SQ problems such as sound-stage depth). But I can't do this with portable devices, as it draws power from the 'phone. That said it's less of an issue with portable electronics as they are less power hungry and create less noise.

So what are the best USB cables? Firstly, be careful. A lot of audiophile USB cables actually increase RF noise and make it sound brighter, and superficially impressive - but this is just distortion brightening things up. Go for USB cables that have ferrites in the cable is a good idea - it may also solve any RF issues from the mobile that you may have too.

 

[mmerrill99]

So you believe that USB cables only deliver bits

No, for some of my applications, the USB cable provides both a 5V DC with up to 500mA power and data to my DACs.  I don't hear any noise on the analog outputs of my DACs when using different computers as sources.  I have not taken any direct measurements, though, as there does not appear to be any problems that would warrant further investigation.  I've used many different USB cables from 1-6 feet in length, though none cost more that $20.  

It also connects the two device grounds, which you failed to mention.

 

[sonitus mirus]

It also connects the two device grounds, which you failed to mention.

 
Yes, but a few of my USB DACs have multiple ways to lift the ground to isolate the DAC from the source, so I don't care too much about that in my applications.  Though, when there is audible ground loop noise, a cable swap will not fix the problem in my experience.  So the cable, be it $5 or $500, would not make a noticeable difference, and the solution to that problem would not be to use a different cable, but to correct the source of the noise or isolate it.  Any cable will transfer that noise, and with a 3 foot length, the math doesn't support a significant variance in dB, unless we start getting into some crazy physical properties with the cables.

 

[mmerrill99]

It also connects the two device grounds, which you failed to mention.

Yes, but a few of my USB DACs have multiple ways to lift the ground to isolate the DAC from the source, so I don't care too much about that in my applications.  Though, when there is audible ground loop noise, a cable swap will not fix the problem in my experience.  So the cable, be it $5 or $500, would not make a noticeable difference, and the solution to that problem would not be to use a different cable, but to correct the source of the noise or isolate it.  Any cable will transfer that noise, and with a 3 foot length, the math doesn't support a significant variance in dB, unless we start getting into some crazy physical properties with the cables.

Well a cable with better RF shielding will have some effect on the composition of the noise on your cable as will different shielding. Whether this will make an audible difference depends on too many variables to predict. So, I believe you are wrong in stating that "any cable will transfer that noise" but I don't believe that you need to spend any more than $20 to make a change.

 

[sonitus mirus]

Well a cable with better RF shielding will have some effect on the composition of the noise on your cable as will different shielding. Whether this will make an audible difference depends on too many variables to predict. So, I believe you are wrong in stating that "any cable will transfer that noise" but I don't believe that you need to spend any more than $20 to make a change.

 
Agreed.  I have a tendency to oversimplify things too much.

 

[mmerrill99]

It also connects the two device grounds, which you failed to mention.

Yes, but a few of my USB DACs have multiple ways to lift the ground to isolate the DAC from the source, so I don't care too much about that in my applications. ....

I missed this in your last post - can you tell us about these ground lift mechanisms - are you talking about the iFi DACs?

 

[sonitus mirus]

I missed this in your last post - can you tell us about these ground lift mechanisms - are you talking about the iFi DACs?

 
I've been using USB DI boxes for my speakers with balanced outputs, which usually allow the ground to be lifted to eliminate ground loop hums.  I haven't had to resort to using the internal transformers, which drastically reduce the FR.
 
http://www.radialeng.com/usbpro-specs.php