RCA analog and digital difference

[Deleted member 89782]

Why not keep it simple?

Digital coax RCA cables work for digital SPDIF and analoge RCA.

Normal RCA cabels work for analoge and may work for digital SPDIF connections; mostly they will.

End of the story.

 

[Kal Rubinson]

Now I'm curious: what exactly is a "75 Ohm connector"? Or a "110 Ohm connector" in case of AES/EBU cables?

A relatively common 75ohm connector is a BNC. (Specifically, a 75ohm BNC as they also come in other impedance matches.)
A relatively common 110ohm connector is XLR. There is an AES specification for this.

 

[Ropeburn]

A relatively common 75ohm connector is a BNC. (Specifically, a 75ohm BNC as they also come in other impedance matches.)
A relatively common 110ohm connector is XLR. There is an AES specification for this.

But really, what exactly has the connector itself got to do with all this electrically? I find it hard to understand what's the difference, other than established use for different types of cables and connections.

 

[antcollinet]

But really, what exactly has the connector itself got to do with all this electrically? I find it hard to understand what's the difference, other than established use for different types of cables and connections.

In RF applications it can be critical. For digital - which might be some 10s of MHz it is important on longer runs.

A good analogy is an optical fibre. The refractive index of the glass is equivalent to the characterstic impdance. At the end of the fibre - when the light emerges into the air - the air has a different refractive index - and you will get reflections at the interface (some of the light will reflect off the cut end of the fibre and go back the other way interfering with the signal.

Similarly take a 75 ohm characteristic impedance cable - the electrical waveform will go happily along it - until it meets a disocntinuity - such as a connector that is a different CI. At that point there will be reflections of the signal back down the cable.

 

[Deleted member 89782]

In RF applications it can be critical. For digital - which might be some 10s of MHz it is important on longer runs.

A good analogy is an optical fibre. The refractive index of the glass is equivalent to the characterstic impdance. At the end of the fibre - when the light emerges into the air - the air has a different refractive index - and you will get reflections at the interface (some of the light will reflect off the cut end of the fibre and go back the other way interfering with the signal.

Similarly take a 75 ohm characteristic impedance cable - the electrical waveform will go happily along it - until it meets a disocntinuity - such as a connector that is a different CI. At that point there will be reflections of the signal back down the cable.

Understood, or so I believe.
But why would an RCA plug be any different than a BNC connector in this respect?

 

[Ropeburn]

In RF applications it can be critical. For digital - which might be some 10s of MHz it is important on longer runs.

A good analogy is an optical fibre. The refractive index of the glass is equivalent to the characterstic impdance. At the end of the fibre - when the light emerges into the air - the air has a different refractive index - and you will get reflections at the interface (some of the light will reflect off the cut end of the fibre and go back the other way interfering with the signal.

Similarly take a 75 ohm characteristic impedance cable - the electrical waveform will go happily along it - until it meets a disocntinuity - such as a connector that is a different CI. At that point there will be reflections of the signal back down the cable.

The optical reflection analogy seems good and I'm fairly familiar with it - two transmission mediums with severely different refraction indexes like air and glass/plastic cause reflections on their common surface, just like air and water do -, but I still don't see how it translates to purely electrical behaviour.

Isn't an electric cable plus connector much more continuous, because the "conductivity" difference between, say, a copper conductor and the plug itself is much smaller? Perhaps I'm thinking too much in LF terms here where that's indeed the case, and don't know enough about RF for it to be intuitive to me.

 

[mhardy6647]

Unfortunately they look much the same although there are differences if you know what to look for.

could I just mention that many coaxial cables have their characteristic impedance printed on them.
Here's a pretty random example via google*

(source: https://shop.klotz-ais.com/c-59u1t.html)

e.g., I think this one is:

• Probably RG-59 coax
• Probably 75 ohm

____________
* I could've gone upstairs and snapped a photo or two... but this was easier and I am preternaturally lazy.

 

[Devnull]

You can think of a cable as in infinite number of RLC connections. The connector might have the same R as the cable but it's the LC components that cause the discontinuity at RF frequencies in the connector. Don't forget that besides the connector on the cable the signal also goes through the connector in the device and also has RLC discontinuities where it attaches to the circuit board.

 

[mhardy6647]

It's amazing we can do anything.
(which actually underscores the close enough counterpoint running through this thread so far)

90-plus year old, and somewhat exotic, Blaw-Knox-built radio tower of WFEA in Merrimack, NH.

Forgotten New Hampshire | The WFEA radio tower in Merrimack | Facebook

The WFEA radio tower in Merrimack. Built in 1932 it’s one of just a few of these types of antennas left in the world. It’s called a Blaw-Knox tower built and designed and patented by that company...

www.facebook.com

The whole tower floats about ground (electrically speaking) -- whether that's DC ground, RF ground, or both... I dunno.

 

[Kal Rubinson]

But why would an RCA plug be any different than a BNC connector in this respect?

For the same reason that there are BNC plugs with different impedance specs: They are structured differently.

 

[Hayabusa]

coax cables only get their characteristic impedance above a certain frequency ( like above 1Mhz or so).
So for analog no discussion, but for digital SPDIF it can matter as the signal can have frequency components above 1Mhz (can go above 10Mhz even)
A wrong impedance will distort the digital waveform and could lead to loss of sync or bit errors.

 

[Devnull]

coax cables only get their characteristic impedance above a certain frequency ( like above 1Mhz or so).
So for analog no discussion, but for digital SPDIF it can matter as the signal can have frequency components above 1Mhz (can go above 10Mhz even)
A wrong impedance will distort the digital waveform and could lead to loss of sync or bit errors.

Even up into the hundreds of MHz the cable's characteristic impedance has a huge tolerance for mismatch. The largest factor impacting signal quality is the termination impedance. As an example USB 2.0 is, IIRC, a 240MHz data stream. It's specd at 90 ohms +- 20 ohms. Hardware should be designed well enough to handle, undershoot, overshoot, ringing, etc. , Obviously within reason. Once you start pushing things your textbook square wave starts looking like a sine wave and that's when EVERYTHING makes a difference, even down to via locations between board layers.

 

[solderdude]

There is a solution for each audiophile problem... gold plated of course.

And indeed... the biggest issue is load impedance.
Now... if I wanted to preserve the actual waveform as exact as possible for say ... waveform analysis (oscilloscope/analyzing) the BNC is the way to go, due to their construction,
For digital audio the exact preservation of the electrical waveform is not really needed, some ringing reflections are allowed.
This has to do with the way data is retrieved (not determined at the edges where most 'problems' are.
How much and type of 'distortion' of the digital waveform is allowed depends on the used receiver and its immunity to 'distorted input'.

 

[antcollinet]

Understood, or so I believe.
But why would an RCA plug be any different than a BNC connector in this respect?

Isn't an electric cable plus connector much more continuous, because the "conductivity" difference between, say, a copper conductor and the plug itself is much smaller?

As I said above - characteristic impedance has nothing to do with the conductivity of the material, but is to do with the mechanical construction (relative dimensions of inner and outer conductors) and the insulation materials used - As stated by..

The inner and outer diameters and insulation materials of a BNC are closely controlled to provide a 75 ohm impedance. There are also 50 ohm BNCs used in RF work. The two will sort of interconnect, but it's considered bad practice. Unfortunately they look much the same although there are differences if you know what to look for.

I once worked in a video facility where it was a sacking offence to bring a 50 ohm BNC onto the premises!

RCA plugs and sockets have no particular impedance although as digital audio isn't very critical, they all pretty much work.

S

But also as stated above - at digital audio frequencies these mismatchs caused by the connector are pretty insignificant.



(This article is much more approachable than the wikipedia one)

Characteristic Impedance | Transmission Lines | Electronics Textbook

Read about Characteristic Impedance (Transmission Lines) in our free Electronics Textbook

www.allaboutcircuits.com

 

[wwenze]

But really, what exactly has the connector itself got to do with all this electrically? I find it hard to understand what's the difference, other than established use for different types of cables and connections.

For example:

As I said above - characteristic impedance has nothing to do with the conductivity of the material, but is to do with the mechanical construction (relative dimensions of inner and outer conductors) and the insulation materials used

Ermmmmmmmm...... nyeh

Mechanical construction and dielectric change the conductivity which is precisely why they affect the characteristic impedance.

- Characteristic impedance is how much current flows into the cable per voltage i.e. same amplitude as resistance i.e. Ohm's Law during the short timeframe after voltage is applied but before reflection comes back from the other end of the cable

 

[antcollinet]

For example:

View attachment 480649



Ermmmmmmmm...... nyeh

Mechanical construction and dielectric change the conductivity which is precisely why they affect the characteristic impedance.

- Characteristic impedance is how much current flows into the cable per voltage i.e. same amplitude as resistance i.e. Ohm's Law during the short timeframe after voltage is applied but before reflection comes back from the other end of the cable

In your pictures above - does the conductivity of the metal change? That is the material being discussed.


Characteristic impedance is related to the distributed capacitance and inductance - that is what is affected by the mechanical construction and dielectrics - not the conductivity of the material. And is given by the ratio of the current and voltage wave front.

The dielectric used impacts the permittivity, not conductivity - or at least any conductivity difference (leakage) is not significant for characteristic impedance.

 

[sam_adams]

This subject has been discussed far too many times in the past here. There is nothing to be gained by rehashing any of this in this current discussion. This has now become just another 'flypaper' discussion where the OP has

and the rest of us are

@RickS, it's time to turn off the lights; don't you think?

 

[RickS]

This subject has been discussed far too many times in the past here. There is nothing to be gained by rehashing any of this in this current discussion. This has now become just another 'flypaper' discussion where the OP has

View attachment 480784

and the rest of us are

View attachment 480785

@RickS, it's time to turn off the lights; don't you think?

Seems likely to get merged. Have to find the right thread though…

 

[lini]

Actually WBT makes RCA/Cinch plugs and jacks with 75 Ohm RF/HF impedance - e.g. the WBT-0110 plug and the WBT-0210 jack. Unfortunately not exactly cheap, as usual for WBT... *sigh*

Digital coax RCA cables work for digital SPDIF and analoge RCA.

Might not necessairly be the best option for analogue use, though. I.e., for one thing cables for electrical SPDIF typically aren't twin cables, but might as well also be made of cable types, that not necessarily sport copper conducturs (like for example MIL-spec RG cable types, many of which have silver-plated, copper-clad steel inner conductors, which one might deem less ideal for analogue audio).

That being said, quite a few cable manufacturers produce RG cable variants, that deviate from the usual MIL-spec versions. For example Van Damme makes an RG179 variant with foam PE dielectric, PVC jacket and OFC copper conductors for both the inner conductor and the shield braid (see there: https://www.van-damme.com/vandamme_product/van-damme-miniature-75-ohm-coax). And Habia makes a small range of RG cable variants, which unlike the regular MIL-spec variants, that they also offer, sport silver-plated copper rather than silver-plated copper-clad steel inner conductors (see there: https://www.habia.com/globalassets/document-downloads/rg-coaxial/coaxials---rg2.pdf - as well as there for the double-shielded variants: https://www.habia.com/globalassets/document-downloads/rg-coaxial/coaxials---rgd2.pdf).

Greetings from Munich!

Manfred / lini

 

[Speedskater]

For a good analog RCA coax cable, in modern interconnect systems (low output impedance, high input impedance) the central conductor can be:
Made of Gold, Gold plated Copper, Silver, Silver plated Copper, hi-fi Copper, Copper, CCA (Copper clad Aluminum), CCS (Copper clad Steel).
It can be any reasonable AWG. It can be stranded or solid.
They will all sound and measure the same.