|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
100.40.104.66
In Reply to: RE: Why coaxials are good for digital and so little popular for analog signals ? posted by beppe61 on March 22, 2014 at 00:20:22
An SPDIF signal's information load consists of the timing of its "On-and-Off" pulses of electrical energy.
Induced reflections from impedance mismatches can blur the crispness of the starts and stops of the signals, because no physical system has infinite rise time. The SPDIF signal is not continuous; the data are encoded in the pulsing on and off of the signal. Whereas in contrast, an analog signal is continuous, and its information load consists of the frequencies and relative levels of its continuous signals.
Look at it this way: If the digital information in an SPDIF signal were being transmitted by the pulses on and off of a signal light, whether the light was on or off would be more important than the exact color of the light. That is because SPDIF is a binary signal--it has only two meaningful states; the rest is noise, which constant impedance is designed to minimize. Whereas an analog signal has hundreds of thousands of potential meaningful states-- 20 Hz to 20,000 Hz times the number of tone colors (harmonics) times the number of volume levels.
What makes coaxial good for digital--its constant impedance, because its "characteristic impedance" is the resistance between the core and shield provided by the insulation--makes it not so good for analog, because of the "skin effect," which means that within the cross-section a solid conductor, the electromagnetic flux propagates at varying speeds, with the signal tending to stay near the skin. This is believed to lead to a smearing of analog arrival times. That is the reason for Litz wire construction and construction using wires of different gauges for analog signals.
Yes, a coax cable will carry an analog signal and in some cases it will sound OK. However, once you get into the deeper end of the pool, the consensus of designers is that continuous analog music signals require special care and handling compared to the on/off of digital signals.
Years ago there was a brief vogue for using solid electrical-power wire for loudspeaker cables. One of the many audio "developments" that were, "one step forward, two steps back."
I think it is relevant to remember that coaxial cable was invented 134 years ago, for the telegraph industry, which used a DC signal--on and off. Telegraphs were a form of digital communication. And also today, coax is mostly used for radio and television frequency applications, that is, very high frequencies far removed from the audio band.
The above is all my own opinions. I do not claim to be an engineer, just a music lover with decades of on-the-job training.
I attach a link to George Cardas' historical review of cable technology, starting with open-wire telegraph systems.
JM
Follow Ups:
Hi and sorry it is me again.
You state " Yes, a coax cable will carry an analog signal and in some cases it will sound OK "
Could you tell me one of this cases are you aware of ?
I would really want to try a coax for analog duties. Really.
And appreciate the suggestion sincerely.
Ok for me would be more than enough.
Actually i had in the past a IC from Audioquest with a green jacket that got lost in a removal. I remeber it has a copper solid wire for + and the shield for -.
I do not remember the exact model unfortunately.
Thanks a lot indeed.
Kind regards,
bg
Edits: 03/26/14 03/26/14
The fundamental function of a 75 ohm digital signal cable is not unlike that of a 75 ohm RF video cable. If folks keep that in mind, the difference between the requirements of a digital cable and a line-level analog cable may be better understood.
The most compelling reason to use a twisted pair or more complicated designs rather than a coaxial cable for analog applications is the ability for the cable to be unshielded, whereas a coaxial cable is shielded by nature. Many audiophiles find that shielding can do more harm than good for analog use.
However, there are coaxial cables that sound excellent when implemented as shielded analog interconnects rather than a shielded twisted pair, et al.
A new kid on the block is the new OCC copper version VH Audio Pulsar II coaxial cable for either digital or analog applications. I'm currently evaluating a stellar sounding DIY VH Audio Pulsar II OCC digital coaxial cable placed in my computer audio system. A new pair of VH Audio Pulsar II OCC analog interconnects are next in line on my DIY to-do list for potential use in my main audio system. I'll post a report sometime down the road.
Hi and thank you very much for your always valuable advice.
You say "Many audiophiles find that shielding can do more harm than good for analog use.
However, there are coaxial cables that sound excellent when implemented as shielded analog interconnects rather than a shielded twisted pair, et al."Could it be the case that for analog use to make the signal going through the shield is a bad design choice ?
In the sense that a shield is ok but connected just to one side for only shielding purposes
and then using the twisted wires for transfer the + and - signals ?
I wonder if the good sounding analog cables work like that.
I do not think that the shield is connected on both side
Great cable the one depicted.
Thanks again.
Kind regards,
bg
Edits: 03/23/14
Hello and thanks for the valuable reply
If I understand well for a good digital signal transmission the impedance must be kept constant along the wire.
And this is relatively easy to achieve and i think maybe can even be measured.
Instead for analog signals the story is more complex with different factors influencing the signal transfer.
And no applicable measurement is available.
I hope i have understood correctly.
Thanks again,
Kind regards,
bg
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: