Jump to content

Sign in to follow this  
jolon

Why digital seems to be affected by power and cables

Recommended Posts

Interesting... so why is balanced so popular for domestic HiFi when low noise unbalanced/RCA design (which is less complicated) is superior?

Share this post


Link to post
Share on other sites

That is an interesting question.  The only reason is to reduce other noise sources.  Balanced is immune to hum loops for example, although I have never had that problem with my Audio Research LS3 preamp, which is unbalanced.  I guess I got "lucky" with my cabling and power layout.

 

Pro users go balanced because of the complexity of the connections - mixing desks are serious bits of kit with a rats nest of wiring and often need to drive loooong cables right next to lighting control looms.  They need to wire these fast before a live gig and know they are not going to have a problem.

 

I run active crossover speakers which have four power amplifiers for each speaker.  I built that lot myself, and went balanced - but simply because with eight active crossover outputs connected to two four-amplifier chassis the possibilities for hum issues was significant.  But other builders of these speakers use unbalanced links and don't seem to have had a problem.

 

This is the wiring - colour coded so I have to try particularly hard to shove the bass signal up the tweeter.  I used the Neutrik XLR's that incorporate RFI filtration on pin 1 to the connector shell - the RFI thing again.

 

post-153047-0-51640600-1449826761_thumb.

Edited by CraigS

Share this post


Link to post
Share on other sites

Interesting... so why is balanced so popular for domestic HiFi when low noise unbalanced/RCA design (which is less complicated) is superior?

 

I wouldn't say that "balanced is so popular" for domestic hifi - I would suggest most components use RCAs ... but some mfrs go balanced for marketing reasons, because some domestic hifi buyers think this is 'sexy' ... and so they buy it.

 

Andy

Share this post


Link to post
Share on other sites

That is an interesting question... I guess I got "lucky" with my cabling and power layout...

No, that is important and it would appear that you did a good job here (although maybe lucky after a second look at the pic :) ). My friend was always careful in that area as am I :)

 

Even worse than RFI (or at least equal) is EMI/EMF effects!

Edited by Ping

Share this post


Link to post
Share on other sites


Also the Philips I2S specification does not define cable impedance, load impedance or whether galvanic isolation is recommended or necessary.  At least they made a better fist of their (and Sony's) SPDIF definition, imperfect though the link is.

 

:thumb:   They just expect the designer to work out an appropriate termination for the specific situation.

Share this post


Link to post
Share on other sites

:thumb:   They just expect the designer to work out an appropriate termination for the specific situation.

 

As a designer that would make me steer well clear of I2S   ;)

Edited by CraigS

Share this post


Link to post
Share on other sites

But if you had a clean sheet to design the best possible digital data link, what would it be?

Already been done.   Master clock is in the DAC and it's sent backwards (in various forms ) to the sources.  Differential signals everywhere.  Nothing that hasn't been around since CTS was put on the first DB25 connector way back when.

Edited by thoglette

Share this post


Link to post
Share on other sites

Ah well - at least it was a good idea!  Too bad it wasn't original.

Share this post


Link to post
Share on other sites


Ah well - at least it was a good idea!  Too bad it wasn't original.

CraigS, does that mean you have no more to say...

 

what a fascinating debate. (rolls up sleeves)

dr_carl, have you now rolled down your sleeves...

Share this post


Link to post
Share on other sites

i said my bit

Share this post


Link to post
Share on other sites

Maybe the thread has just run its course

Yes, it seems so :)

Share this post


Link to post
Share on other sites


Reflections can be reduced by resistance. madscientist-audio suggests a 37Ohm cable in lieu of 75Ohm. which he offers. Interesting concept. no engineer here or Lab guy just open eyes.

   

 

charles

Share this post


Link to post
Share on other sites

Don't confuse a cable's DC resistance end to end with characteristic impedance.  I have no idea what the characteristic impedance of his cable is - it might be 75 ohms and it might not.  But having a cable with really significant series resistance basically slugs the rise time (look up TDR of lossy cables and you'll see), and 37 ohms of series resistance will slug it rather a lot.  Anyone who knows the first thing about signal integrity (like Agilent, LeCroy and Tektronix) tries very hard indeed to minimise transmission line resistance.  If I feel the urge to do the sums for a 37 ohm resistive cable I'll let you know the result.

 

Like many of the purveyors of audio science fiction, you'll note that he quotes no real numbers or measurements, and shows no signal waveforms (it is easy to do with a 75 ohm power splitter [minicircuits], some cable, a 75 to 50 ohm pad [minicircuits again] and a 50 ohm load and an oscilloscope with 500MHz bandwidth).  So why does he not to that?  Because the signal will look like a dog's dinner.

 

Slugging the rise time opens up the link to jitter - precisely the opposite of what the guy claims.

Share this post


Link to post
Share on other sites

Reflections can be reduced by resistance. madscientist-audio suggests a 37Ohm cable in lieu of 75Ohm. which he offers. Interesting concept. no engineer here or Lab guy just open eyes.

   

charles

Do you have a link so that we can have a read :)

Share this post


Link to post
Share on other sites


http://blackdiscus.blogspot.co.nz/2015/11/heretical-digital-cable.html

 

Resistance soaks up the reflections. This is probably more important than skin effect. The resistance of the conductor, being of similar magnitude to the 75-ohm loading means that reflections are not going to be able to do much damage - they will be turned into heat.

 

The Heretical Digital Cable has a resistance of about 37 ohms - half the 75 ohm characteristic impedance. But all SPDIF inputs are terminated with 75 ohms. This means that a reflection that reflects off the DAC end of the cable/plug will travel towards the source where it can interfere with the data. But the resistance will tend to turn the energy into heat, so reflections are dissipated very quickly.

It appears that he has confused impedance with resistance although carbon fibre does have a much higher DC resistance when compared to copper.  It appears that he is assuming the the reflections are low energy and will dissipate rapidly to 0 volts. I would assume that if the reflections travel towards the source that they would interfere with the signal coming from the source and ultimately change it (not good).    It what he says does work then it could simply be a low impedance to higher impedance matching that is occurring (by chance).   I would think that this may not be the situation with other/all SPDIF connections!    I would think that ultimately speed combined with impedance matching is the real solution to the issue :)

Share this post


Link to post
Share on other sites

 

That is a more informative link.  There is the key phrase "The crucial point is this : If you have a 75-ohm transmission line (like a digital cable) and you have some part of the system that is NOT 75-ohms, then you get signal reflections."  His solution to that is to use a resistive cored cable to absorb the reflections.

 

So - if you do have a true 75 ohm environment (see my earlier posts) you just use a high quality, but low cost 75 ohm cable.  And because audio designers always get this wrong, I have replaced all the SPDIF inputs and outputs with designs that offer 75 ohms up to 2GHz.

 

The problem with using a resistive core is that it produces an effect called "dribble up".  Every tiny part of the cable's length produces a small reflection, because of the fundamentals of how a lossy transmission line work.  The net effect is that after a small amplitude fast rise, the rest of the signal increases very slowly.  The difficulty in applying the conventional lossy transmission analysis is that it supposes that the loss occurs through skin effect, which is frequency dependent.  Madscientist correctly says that highly resistive materials (like his carbon fibre core) do not suffer skin effect.  However, dribble up will still occur; it would be interesting to actually *measure* one of his cables.

 

There are some examples of dribble up here https://kh6htv.files.wordpress.com/2015/11/an-03a-pulse-meas.pdf although these are for short lengths of conventional coax, so the timescale is in tens of picoseconds, rather than the tens nanoseconds I would expect with the carbon fibre cored coax.

Share this post


Link to post
Share on other sites

If SPDIF, just like XLR, circuitry is not as good as it should be (ie true 75ohm impedance), which appears to be the usual scenario, then what is a good SPDIF compromise with what is already existing out there in HiFi Land?   

Is "dribbling" up/along better than other forms of noise/compromise/etc?

 

If SPDIF and XLR have compromise issues then I guess most USB circuits also have compromise issues?

Edited by Ping

Share this post


Link to post
Share on other sites

This illustrates precisely what I have been talking about regarding rise time and pulse fidelity.  Although it is about SDI signals for digital video, the principles are identical SPDIF and other serial links in which the clock is embedded in the data.  I'll extract one relevant comment and leave you to read the full document yourselves:

 

Deterministic jitter
==============
A wide range of sources can introduce deterministic jitter
into an SDI signal. For example:
Noise in a switching power supply can introduce periodic
deterministic jitter.
The frequency response of cables or devices can introduce
data-dependent jitter that is correlated to the bit
sequence in the SDI signal
Differences in the rise and fall times of transition can
introduce duty-cycle dependent jitter
 

Share this post


Link to post
Share on other sites

Thanks for the correction about impedance and Ohms. The cable from Mad scientist just sounds terrific. Replaced a JPS labs and Belkin silver.

 

charles

Share this post


Link to post
Share on other sites

Why should this discussion change your mind about the Mad Scientist SPDIF?    The measurements cannot change your ears?    This is only technical babble and it would appear that other measurements, that are not currently done or known, are at least as important, and perhaps more important, than the usual 75ohm impedance measurement.    It also appears that AES/EBU, USB, Ethernet, etc, will suffer from similar issues :)

 

It also appears that the JPS labs and Belkin silver SPDIFs are not very good and the reviews and posts about them are exaggerated or not true :)   

Share this post


Link to post
Share on other sites

Well there you go - why design anything in audio based on electronic principles?  Why not randomly just connect electronic bits and pieces until it sounds right?

 

But hey - what do I know?  1st class electronics degree and Phd in laser physics, and ex CTO of Wharfedale.

 

Over to you guys - I'm signing out of this thread.

Share this post


Link to post
Share on other sites

Interesting, perhaps you might know why the Mad Scientist SPDIF, with an impedance of 37ohms, is working/sounding better than either of the 75ohm JPS labs and Belkin silver SPDIFs?    I would imagine that at least the Belkin silver SPDIF would have a 75ohm impedance?   

 

Perhaps, the JPS labs and Belkin silver SPDIFs are just not good or the reviews and posts are rubbish?   

Perhaps, I guess, rollo's electronics could be rubbish, but that is what rollo and all of us have to work with?

 

Perhaps, you can unravel the enigma? ... but then again with all the current electronics out there, there is no absolute answer or manufactures are not willing to employ the absolute answer/solution :)

Edited by Ping

Share this post


Link to post
Share on other sites

Have you been reading anything I have been saying on this?

 

Definitely signing out

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Classifieds Statistics


    Currently Active Ads

    Total Sales (Since 2018)

    Total Sales Value (Last 14 Days)

    Total Ads Value (Since March 2020)
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...