Digital Coax vs RCA cable

[Gambit 1870]

Hi brains trust. 

 

Is there any audible sonic difference between the two cables. 

I understand the coax has a higher impedance value for better bit rate performance , just curious about the sonics as I'm currently just using and RCA cable between streamer and dac? 

 

Thanks

Edited May 10, 2022 by Gambit 1870

[ZGojira]

Without setting off another heated debate.....I think a good way for you to find out is to try using another RCA cable (assuming you have one) in place of your current one, and see if you can hear a difference.

 

If you do hear a difference, then continue to investigate into whether you want a coax cable.

 

 

Edited May 10, 2022 by ZGojira

[Guest]

It depends on the source, destination and cable.  In theory, all are equal and there should be no difference...

 

Best to try for yourself on your setup.

[Ittaku]

It very much depends on the implementation at both ends. In theory optical should be virtually lossless, but most optical is limited by the quality of the sender/receiver components and rarely meets the quality of a coaxial connection since it's converted from electrical to optical and then back to electrical again anyway.

[Gambit 1870]

Thanks guys, I don't have a "coax" to compare hence the question, but sounds like the answer is not likely

[bob_m_54]

Coaxial Cable is designed to have certain impedance characteristics at a particular frequency, and generally used for higher frequencies in the RF spectrum. To achieve this, the cable is made with higher tolerances in dimensions of the centre conductor diameter, dielectric thickness, and the type of dielectric.

 

RCA cables are generally a shielded cable, which is merely an insulated cable (or pair) with a shield around it, and not designed for a specific impedance, and not suitable for operating at high frequencies.

 

As far as digital data is concerned, you may be able to use a shielded cable, but pulse shaping could be degraded if it happens to be too far from the optimum impedance. In practice, it may work OK, or it may not work reliably, especially at longer distances.

 

[aechmea]

I use cables that are designed for the job and conform to the specifications.

 

I use 110ohm AES/EBU cable where I have XLR connections available.

I use 75ohm coax for BNC connections.

Shielded RCA cable is generally 50ohm, so doesn't really meet the specs.  Many people use it, so it's a "maybe".

I don't use Toslink as I have never found a plug/socket combo that seems to mate securely.  Seems sloppy/wobbly to me.

 

The specs are there for a reason so I conform as best I can with the proper cable for the job.

 

All of this assumes that the streamer, DAC and cable manufacturers know what they are doing re specs and their design/parts.

Edited May 10, 2022 by aechmea
Removed "coax" from AES/EBU cable

[muon*]

Only way to know is by trying yourself.

 

BTW some RCA leads are just coax.

[rockeater]

2 hours ago, bob_m_54 said:

Coaxial Cable is designed to have certain impedance characteristics at a particular frequency, and generally used for higher frequencies in the RF spectrum. To achieve this, the cable is made with higher tolerances in dimensions of the centre conductor diameter, dielectric thickness, and the type of dielectric.

 

RCA cables are generally a shielded cable, which is merely an insulated cable (or pair) with a shield around it, and not designed for a specific impedance, and not suitable for operating at high frequencies.

 

As far as digital data is concerned, you may be able to use a shielded cable, but pulse shaping could be degraded if it happens to be too far from the optimum impedance. In practice, it may work OK, or it may not work reliably, especially at longer distances.

All of this is above very true.

 

I fix CD players for a living and also make videos about it.

Accidentally, between the repair of the player and shooting a video, I have found that the SPDIF signal on the oscilloscope has changed; from almost perfect, to rounded squares with overshoots added.

So I started comparing any cable I coiuld find in my workshop not by listening, but by looking at the oscilloscope traces and the differences are real.

As @bob_m_54 said, the ones that are specifically marketed as "digital" cables perform better.

Even better than them was a Belden double-shielded, teflon coated coax for UHF radio use, even though it was 50Ω. (which means it will cause a minimal impedance mismatch, so a 75Ω version  - if it exists - would have been better).

 

And there is a treason for it.

Audio signals are ending at 20kHz, whereas the SPDIF is around 1.5MHz (from memory).

[Ittaku]

75 ohm cables are not that hard to come by, nor expensive, if you know where to look. These usually perform better than many "audiophile" expensive brands which aren't actually 75 ohm.

 

https://www.selby.com.au/brands/origin/75cm-neotech-origin-coax-digital-audio-video-cable-ori7500.html

[Monkeyboi]

Coaxial S/PDIF cables should be 75 ohms impedance and are unbalanced.  Most hi-fi gear uses RCA (phono style) connectors for the electrical S/PDIF link which by virtue of their physical geometry aren't 75 ohms even though some cable manufacturers would lead you to believe otherwise.  Commercial gear uses BNC connectors which BTW come in two typical impedances (75 and 50 ohms and are slightly physically different.)

Ideally you should be using the correct cable for optimum signal transmission, however you can(at a pinch) get away with video cables which should be 75 ohms which is the industry standard impedance for video.  Audio coaxial cables aren't suitable even though they might work, are far from ideal for the job.

 

AES/EBU cables are 110 ohms balanced and have two twisted wires (a pair) in a screened jacket terminated with 3 pin XLR style connectors at each end .  They aren't coaxial cables.

Edited May 10, 2022 by Monkeyboi

[rockeater]

2 hours ago, Monkeyboi said:

however you can (at a pinch) get away with video cables which should be 75 ohms which is the industry standard impedance for video.

This is a good idea for getting a digital cable on the cheap.

Video bandwidth is much higher than audio and they are 75 Ohm so they should be much better than analogue RCAs.

I will try this and will make a video, comparing them to all the other ones I have.

After I finish the current job which is CDP-337ESD. It has an excellent SPDIF output, so it should be ideal for this.

[aussievintage]

A couple of thoughts.  Firstly many cables are coaxial.  Your simple shielded cable used for RCA plugs is usually coaxial. So when people are referring to "coax" here, they are generally meaning a shielded cable constructed to have a known targetted characteristic impedance.

 

Characteristic impedance of a cable is the impedance of a theoretically infinite length of the cable.  However, a shorter length will behave as if infinite, if terminated in the same impedance as the characteristic impedance.  If not, and here's where problems occur, you will get reflections* and lower levels of signal transferred.  So using a simple RCA shielded cable means you cannot be sure it is terminated in the proper impedance because it's characteristic impedance is not really known.

 

* In the days of analogue television, you could get "ghosting" ( a shifted secondary image on the screen) in the picture if proper splitters and terminations (at the correct impedance) was not used.  The ghosting is these reflections moving up and down the badly terminated coax.  Imagine what this might do to digital data if it is bad enough.

 

 

[Ittaku]

Most RCA plugs do not meet the criteria to be 75 ohm termination. However there are some specially designed with crimp terminations to meet 75 ohm. The vast majority though are solded and it's pretty much impossible to meet 75 ohms in that case, if I'm not mistaken.

[bob_m_54]

The thing is, Coaxial Cable was originally designed for RF transmission use, so the dimensional aspects of the construction are design critical, and great care is taken to ensure the dimensions are accurate and consistent throughout the length of the cable. Also the dielectric properties of the insulation between the two conductors (outer and inner) is design critical as well. And note; the outer braid is termed as the outer conductor, not "shield" for Coaxial Cables.

 

 

But a single conductor Shielded cable as used in many standard RCA interconnects, even though the construction is in a coaxial configuration, it isn't made to meet the tight dimensional constraints of Coaxial Cable, and the insulation between the conductor and the shield isn't selected for dielectric properties as much as for insulation strength and durability.

 

As for connectors, RCA connectors, the ones that have solder tabs can't be considered to have a specific impedance, due to the geometric inconsistency of the solder tabs in relation to each other, and the way the cable is attached.

 

 

Whereas the connectors with a centre contact pin attached to the centre conductor, and the outer conductor slips over a ferrule and held in place with an outer crimped ferrule, can be made to be much closer in tolerance, and designed to be close to the desired impedance.

 

[Ittaku]

I should also point out that many advertisements for 75 ohm coaxial cables are worded in such a way as to not clear they mean that ONLY the conductor is 75 ohm, with no reference to whether the whole terminated cable is. You'll see the one I linked above at Selby lists "True 75 ohm" as one of its features, whereas a very similar ad features "75 ohm cable" in its features, which doesn't make it clear if it's just the cable - as in the wire - that's 75 ohm, or the whole construction.

Edited May 11, 2022 by Ittaku

[aussievintage]

13 minutes ago, bob_m_54 said:

But a single conductor Shielded cable as used in many standard RCA interconnects, even though the construction is in a coaxial configuration, it isn't made to meet the tight dimensional constraints of Coaxial Cable, and the insulation between the conductor and the shield isn't selected for dielectric properties as much as for insulation strength and durability.

 

Yes.  If I wasn't clear, audio shielded cable, although it IS coaxial, is not a good choice.

 

14 minutes ago, bob_m_54 said:

Whereas the connectors with a centre contact pin attached to the centre conductor, and the outer conductor slips over a ferrule and held in place with an outer crimped ferrule, can be made to be much closer in tolerance, and designed to be close to the desired impedance.

 

But this is still just a connector.   If plugged into equipment, the circuitry inside forms an important part of a proper "termination" at the characteristic impedance.   The connector is designed to maintain the characteristic impedance of the cable and not be a discontinuity to it, like a kink in a cable will be.  Such things can also cause loss and reflections.  So much so that the distance down a cable to a fault can be measured from the end of the cable using a TDR (time domain reflectometer) which times the reflection and so predicts the distance.

 

A proper terminator for unused cable or ports can look like this (not that they are normally needed in audio situations, but can be in video)

 

 

 

[bob_m_54]

2 minutes ago, aussievintage said:

But this is still just a connector.   If plugged into equipment, the circuitry inside forms an important part of a proper "termination" at the characteristic impedance.   The connector is designed to maintain the characteristic impedance of the cable and not be a discontinuity to it, like a kink in a cable will be.  Such things can also cause loss and reflections.  So much so that the distance down a cable to a fault can be measured from the end of the cable using a TDR (time domain reflectometer) which times the reflection and so predicts the distance.

Yes, I agree.. what i was saying in layman's terms is the soldered connectors are rough as guts as far as what they can do to integral impedance of the cable-connector-interface. And yes, I am very familiar with the TDR both for EM and Optical cables LOL.

 

As an interesting aside. We had an aircraft in maintenance, where the Metal Worker had drilled a 4mm hole into a Coaxial Cable feeding one of the antennas on the aircraft (can't remember which system now) and our ex Aircraft Electrician-Avionics Techs said... No problem, just fill the void in the dielectric with some filler, and do a a braid wrap repair on the outer conductor brad. It's covered in the Repair Manuals under "Shielded Cable Repair Methods"

 

I was, at the time an ex Aircraft Radio Technician-Avionics Tech. It took me a lot of arguing with them and the boss (ex Aircraft Instrument Fitter-Avionics Tech) to convince them that this was not possible, and that the only authorised repair method for the Coaxial Cable was to replace that whole cable segment, which was a major job (in excess of a couple of hundred manhours). In the end they compromised and were given a dispensation from the Engineers (Mechanical Engineers with Electrical sign off capabilities) to cut and join the cable using the proper connectors, till the next major servicing. Just goes to show what can happen when the guy doing the repair doesn't understand the difference between shielded cables and coaxial cables.

[Monkeyboi]

For short lengths of coaxial cable as one would encounter in a domestic hi-fi setup, the impedance of the RCA connector is for all intent and purpose irrelevant.  I seriously doubt that even if the "impedance" of the connector and socket were 75 ohms there are probably no impedance controlled traces on the average PCB to maintain the correct impedance at 75 ohms all the way from the connector to the receiver chip input.  Likewise from the transmitting chip to the connector in the CD / DVD player.

 

Astute readers would have probably noticed that high end and professional DACs use 75 ohm BNC connectors in lieu of RCA for their coaxial S/PDIF connections.  So why is it we so commonly see RCA connectors fulfilling this role?  The simple answer is cost.

[aussievintage]

17 minutes ago, Monkeyboi said:

For short lengths of coaxial cable as one would encounter in a domestic hi-fi setup, the impedance of the RCA connector is for all intent and purpose irrelevant. 

 

I would caution that there will be times when it is relevant.   I have done it for testing,  but...

 

19 minutes ago, Monkeyboi said:

I seriously doubt that even if the "impedance" of the connector and socket were 75 ohms there are probably no impedance controlled traces on the average PCB to maintain the correct impedance at 75 ohms all the way from the connector to the receiver chip input.  Likewise from the transmitting chip to the connector in the CD / DVD player.

 

Like you said,  short runs.   Luckily it's a controlled situation.  Once tested and working, the traces won't change, so will likely continue to work.

 

20 minutes ago, Monkeyboi said:

Astute readers would have probably noticed that high end and professional DACs use 75 ohm BNC connectors in lieu of RCA for their coaxial S/PDIF connections.  So why is it we so commonly see RCA connectors fulfilling this role?  The simple answer is cost.

 

Yep, you can get away with being cheap for short runs.   Till you don't, then the average punter will wonder why.

[Gambit 1870]

Thanks for all the insight guys, actually a bit more in it than I first thought. 

Looks like I'll spend $20 at Selby's and let my ears be the judge. At the worst I'll have a set of Chord Crimson RCA's to offload