DIY SPDIF Cable?

[thommy]

Where is the 75 ohm thing? Is it in the cable? Cos I can see a 75 ohm resistor near the spdif input of my dac which makes me think an spdif cable is just a normal rca lead made of something that is more sympathetic to digital signals (and well shielded).

I have some really nice braided screened PTFE Teflon magic wonderful foo silver wire, expensive shit, came from a lovely chap who worked on government projects and had some left over. Can I bung a pair of Neutrik RCA's on the ends and call it a high end digital transport interconnect?

[thommy]

I guess the test will be comparing it to my Missing Link Dark Art, but would be interested to know what people think.

[Barry]

Where is the 75 ohm thing? Is it in the cable? Cos I can see a 75 ohm resistor near the spdif input of my dac which makes me think an spdif cable is just a normal rca lead made of something that is more sympathetic to digital signals (and well shielded).

I have some really nice braided screened PTFE Teflon magic wonderful foo silver wire, expensive shit, came from a lovely chap who worked on government projects and had some left over. Can I bung a pair of Neutrik RCA's on the ends and call it a high end digital transport interconnect?

The cable itself needs to have a characteristic impedance of 75 Ohm as well for it to be optimum. There are several cables to choose from, and they need not be expensive.

[awkwardbydesign]

If, like me, you wondered what it's all about, this is useful. http://www.bluejeanscable.com/articles/impedance.htm

[thommy]

Oh that's brilliant, nice one!

[Light Dependant Resistor]

If, like me, you wondered what it's all about, this is useful. http://www.bluejeanscable.com/articles/impedance.htm

What designates characteristic impedance of one cable to another is the physical distance
between conductor and the shield. A 50 ohm cable being slightly closer than a 75 ohm cable.
Coaxial cable exists to maintain characteristic impedance over any given length, so if
low impedance like this is needed, it is the natural choice to use.

Having some other cable that is not coax and forcing 75 ohms with a resistor is false, enough said

The benefit of maintaining exact impedance is lost where the strict distance relationship varies
over the source point to load end, what occurs then are reflections and what are called standing
waves arguing with the source, and if you want to get fussy includes circuit board tracks
which will usually only be designed properly in RF amplifiers, but of course there is invitation
to do it properly in all equipment, :)

So the end connectors should then mimic the coax cable to maintain distance relationship
throughout entire length.

I have had many years of employment in broadcasting and technical provision and seen
coax that is over two feet wide, and others that are tiny, they all do the same thing but
at different power capability levels.

A RCA termination for instance is not 75 ohms or 50 ohms , rather a BNC falls somewhere
between 50 and 75 ohms as it is used in both 50 ohm broadcast standard and 75 ohm applications.
It should not be excused to use a RCA but usually is, in favour of ease of connection.
It all comes down to your focus on what physics dictates and how much care is then
provided to get desired results.

The history of coax is great bedtime or other time reading https://en.wikipedia.org/wiki/Coaxial_cable

Cheers / Chris

[thommy]

So cable impedance is inversely proportional to the distance between screen & signal? i.e. The thickness of the insulation around the signal part of the cable affects it's performance.

[thommy]

I still don't understand this impedance figure properly.

If I measure a resistor from either end I get the impedance of the resistor on the screen of my DMM however if I measure the impedance between conductor and shield of a piece of coax I get open circuit on the DMM which makes sense.

So, where can this 75 ohm impedance be measured? Is it the resistance between shield and conductor when the system is fully connected from spdif output chip to the receiver chip before the DAC?

[struth]

You would need a scope or maybe a sig gen and a swr bridge..impedance is not resistance

[Stratmangler]

So, where can this 75 ohm impedance be measured? Is it the resistance between shield and conductor when the system is fully connected from spdif output chip to the receiver chip before the DAC?

It can't be measured with a DMM.
The S/PDIF is a transmission line, and its impedance is its characteristic impedance.
You can see the impedance on an oscilloscope.
The cable should be of negligible impedance (ie effectively no impedance), as will the connectors on the ends of it when measured with a DMM.

[NRG]

Resistance is a DC measurement IE 0hz, like your DMM. Impedance is an AC measurement at some frequency IE: 1Mhz

The use of RCA connectors with SPDIF screws up the impedance matching royally so there is no use in getting all anal about the characteristic impedance of the coax able. As far as I recall an RCA connector has an impedance of around 35~40 Ohms, this gross mismatch with the nominal desired impedance of 75 Ohm causes a step in the impedance and results in a reflection of the signal back to the transmitter whereupon its re-reflected back to the receiver.

With a matched connection and correct cable termination this reflection does not occur or if it does the reflected signal is so low in amplitude that it does not matter.

As the reflection bounces back and forth it loses amplitude and decays due to there being some but not ideal termination of the signal even with the impedance mismatch.

This mismatch can be measured with a TDR or with the use of an oscilloscope and signal generator.

Some will argue it makes no difference at all what you use for SPDIF others will say the complete opposite, if you have RCA connectors IMHO you are better of making sure the cable is double or triple shielded and sticking to cable length multiples of 1.5m ;)

There's loads more to this if you google it....

[thommy]

Thanks for all the replies, this is fascinating!

[Light Dependant Resistor]

Thanks for all the replies, this is fascinating!

Where you get to see these principles really working is with antenna's
where the delivery at a given frequency has current at maximum
at the coax connection, or very middle, and voltage reaching maximum
at the antenna ends. Every good transmitter installation has a standing wave ratio
meter to assess reflected waves. Wet weather can and does dip SWR
( standing wave ratio )

Dipoles are very popular, but the most interesting antenna I think is the discone
as it has very good impedance matching and capability of a very wide bandwidth
Walk past a fire station or police station and you are very likely to see one
as they are not only good for transmitting but also receiving.

Understanding frequency and the speed of radio waves then gives you ability
to design antenna dimensions.

Back to SPDIF , it is a complex digital multiplex harbouring a number of frequencies
that then get translated back from where they started as Data Bitclock and Left Right Clock
I looked into a more direct method to connect digital data in 1997
http://www.enjoythemusic.com/magazine/viewpoint/0401/deficienciesofspdif.htm

Cheers / Chris

[Audio Advent]

I still don't understand this impedance figure properly.

If I measure a resistor from either end I get the impedance of the resistor on the screen of my DMM however if I measure the impedance between conductor and shield of a piece of coax I get open circuit on the DMM which makes sense.

So, where can this 75 ohm impedance be measured? Is it the resistance between shield and conductor when the system is fully connected from spdif output chip to the receiver chip before the DAC?

Think of impedance via a physical analogy..

Impedance is the resistance to oscillation - so some physical objects will be able to oscillate at low frequencies and will resist oscillating at higher frequencies. Similarly some objects will happily oscillate at high frequencies and will resist, mute attempts to make it oscillate at lower frequencies.

Or you can perhaps imagine how easy it is to move something back and forth through the air compared to water compared to treacle - kind of a similar resistance to oscillation you can "feel" .

And so impedance in electronics is a measurement determining how much the conductor will resist the electrical oscillations at different frequencies. For S/Pdif, these are signals at megahertz area, million times a second. There will be all sorts of back emf and capacitance effects at difference frequencies so the make up of a cable and all the conducting surfaces nearby will influence how freely the signal can oscillate and carry charge back and forth along the cable. The 75ohms figure is what is needed at this specific frequency (so you'd have to measure by passing that frequency through the cable).

One area where those analogies don't work is that it is the electrical fields and magnetic fields created by the very signal being transmitted and it's interaction with the cable which creates all the impedance in the first place. Hence if there is no oscillating signal there is no impedance, it measures zero.