IPB

Welcome Guest ( Log In | Register )

Coaxial SPDIF cable?
bennetng
post Nov 25 2012, 15:19
Post #1





Group: Members
Posts: 233
Joined: 22-December 05
Member No.: 26587



We were told only use 75 ohm cable for SPDIF purpose. So what is the impedance of typical analog RCA audio cable?

Another question is I have a 75 ohm TV antenna cable, is it suitable for SPDIF purpose?

Some people said that wrong impedance may not cause problem if the cable is short enough, if it is true, then what cable length will cause problem?

I ask these because I just made a SPDIF cable with two spare RCA plugs and a 75 ohm TV antenna cable (3 feet), I tested it using RMAA and found no problem (+- 0dB freq. response, -146dB noise level, 0% THD and so on) and I can't hear any problem as well. Does it mean that I don't need to spend money to buy a standard cable?

Thanks.
Go to the top of the page
+Quote Post
 
Start new topic
Replies (1 - 22)
Speedskater
post Nov 25 2012, 17:41
Post #2





Group: Members
Posts: 139
Joined: 12-December 07
From: Cleveland, OH
Member No.: 49500



a] The longer the cable the more important the 75 Ohms is. At 3 feet it's not important, at 50 feet it's rather important.
b] But some poorly designed components are sensitive to cable differences and lengths.
c] Typical nice analog audio cables very well may be about 75 Ohms. It's the cheap sweet spot for cable manufacturing.
d] RG-6QS (Quad Shield) is designed for the very high frequencies of cable TV so it's not the best choice for digital or analog audio. Ordinary RG-6 is a better choice.
e] At 3 feet, I suspect almost any cable would work.

(Yes I have references for [b] & [c] )


--------------------
Kevin
Go to the top of the page
+Quote Post
pdq
post Nov 25 2012, 18:00
Post #3





Group: Members
Posts: 3443
Joined: 1-September 05
From: SE Pennsylvania
Member No.: 24233



Analog audio cables may not be closely matched to 75 ohms (they don't need to be), but analog video (composite video) cables with RCA connectors (the yellow ones) are an excellent choice for SPDIF digital.

As was said, short cables (a few feet or less) should work with just about any coaxial cable.
Go to the top of the page
+Quote Post
mzil
post Nov 25 2012, 18:14
Post #4





Group: Members
Posts: 710
Joined: 5-August 07
Member No.: 45913



Besides the wire itself, the plug used on the end partly dictates the overall impedance based on its geometry, including the center core's diameter and distance (the ratio) from the outer shield. F-pin and BNC connectors would technically be the proper termination to maintain the 75 Ohm rating that S/PDIF is looking for, however in real world use, and looking at the jack panel on the back of most consumer gear, RCA works just fine, at least for the short distances we might use in a home.

Properly terminating coaxial wire such as RG59 or RG6 (often used for TV/cable/satellite wiring) should work just fine with RCA plugs for your application, in fact it might be even better (in practice) than commercially purchased RCA wires that claim they are "75 Ohm". [Non truly are, the RCA's geometry simply wont allow it, as they were never designed from the get go to be 75 Ohm, but they are close enough for most situations, so they work.]

The wire you have prepared should be fine.

I don't think there is an answer to your question "At what distance should I start to worry" because it is very gear dependent. In one situation a 4 meter wire might have problems, yet in another situation a 6 M wire works fine. If all your wires stay within the same rack and don't exceed 3 to 4 meters tops, I doubt you'll have any problem, with most gear. If you have to run the wire across the room or to other rooms of your house, there could be issues.
Go to the top of the page
+Quote Post
Glenn Gundlach
post Nov 25 2012, 18:17
Post #5





Group: Members
Posts: 372
Joined: 19-April 08
From: LA
Member No.: 52914



QUOTE (bennetng @ Nov 25 2012, 06:19) *
We were told only use 75 ohm cable for SPDIF purpose. So what is the impedance of typical analog RCA audio cable?

Another question is I have a 75 ohm TV antenna cable, is it suitable for SPDIF purpose?

Some people said that wrong impedance may not cause problem if the cable is short enough, if it is true, then what cable length will cause problem?

I ask these because I just made a SPDIF cable with two spare RCA plugs and a 75 ohm TV antenna cable (3 feet), I tested it using RMAA and found no problem (+- 0dB freq. response, -146dB noise level, 0% THD and so on) and I can't hear any problem as well. Does it mean that I don't need to spend money to buy a standard cable?

Thanks.


I just tried searching "transmission line impedance mismatch" and got a lot of pages including 2 Wiki pages with a lot of math that will tell the average person nothing useful. What happens with mismatched lines is the signal doesn't get absorbed on either end of the cable and reflects back and forth on the line. This shows up as pulse distortions depending on the length of the cable vs the length of the pulses involved. Get enough distortions and you may reach the point where the digital information is _sometimes_ wrong. More distortions and more data lost.

I did find a few pages to show the effect and this one is quite good. While the pulse times are faster than SPDIF, the principle is the same.

http://www.avtechpulse.com/appnote/techbrief12/

The simplest way to avoid problems is to use proper transmission line and connectors.

An important thing to remember is there is no such thing as a 'digital' signal on a cable. What it is is an ANALOG _representation_ of a digital signal. All the problems with analog signals on a cable (loss, HF roll off, reflections, etc) apply to 'digital' signals. Keep the analog signals clean and you'll have no problems.

G

This post has been edited by Glenn Gundlach: Nov 25 2012, 18:30
Go to the top of the page
+Quote Post
greynol
post Nov 25 2012, 18:27
Post #6





Group: Super Moderator
Posts: 10256
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Speedskater @ Nov 25 2012, 08:41) *
I have references for [b] & [c]

I'd be interested in a reference for [d].


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
Soap
post Nov 25 2012, 18:36
Post #7





Group: Members
Posts: 1017
Joined: 19-November 06
Member No.: 37767



QUOTE (Speedskater @ Nov 25 2012, 11:41) *
d] RG-6QS (Quad Shield) is designed for the very high frequencies of cable TV so it's not the best choice for digital or analog audio. Ordinary RG-6 is a better choice.


What, besides cost and flexibility, is the performance downside of RG-6QS in this application?


--------------------
Creature of habit.
Go to the top of the page
+Quote Post
bennetng
post Nov 25 2012, 18:45
Post #8





Group: Members
Posts: 233
Joined: 22-December 05
Member No.: 26587



Thanks all. Although I don't have enough knowledge to understand all of the replies it seems that my cable should just work fine.

I think my cable is RG-6 after a photo search.
Go to the top of the page
+Quote Post
washu
post Nov 25 2012, 19:03
Post #9





Group: Members
Posts: 136
Joined: 16-February 03
From: Ottawa
Member No.: 5032



I've found SPDIF rather resilient when run over cables that are not to spec. I've got a 30' run that is made of a standard 1/8" phono jack extension cable (because it's what I had) with adapters on either end. The far end has a splitter to feed it into an old set of cambridge soundworks 4.1 speakers that needs separate inputs for the front and back pairs. A very hackish solution and definitely not in spec, but it works fine.
Go to the top of the page
+Quote Post
greynol
post Nov 25 2012, 19:08
Post #10





Group: Super Moderator
Posts: 10256
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Are you sure your speakers aren't being fed by a TTL signal?


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
pdq
post Nov 25 2012, 21:52
Post #11





Group: Members
Posts: 3443
Joined: 1-September 05
From: SE Pennsylvania
Member No.: 24233



QUOTE (mzil @ Nov 25 2012, 12:14) *
F-pin and BNC connectors would technically be the proper termination to maintain the 75 Ohm rating that S/PDIF is looking for

I thought that BNC connectors were matched to 50 ohms?
Go to the top of the page
+Quote Post
mzil
post Nov 25 2012, 22:49
Post #12





Group: Members
Posts: 710
Joined: 5-August 07
Member No.: 45913



Both exist:

http://en.wikipedia.org/wiki/BNC_connector
---

QUOTE
I think my cable is RG-6 after a photo search.

RG6 and RG59 look the same in photos, but RG6 is slightly larger and stiffer, which makes it harder to work with in a tight stereo cabinet. If the center core wire diameter is around 1.02 mm, then you have RG6, if it is .64 mm then you have RG59. Both should work well but be careful not to bend them too tightly; in best practice never more than 6 inch diameter circles should ever be attempted. Many people understand the danger of snapping the inner core, but what a lot of people don't realize is you also potentially can compromise the performance of the wire if it is bent more tightly than this, even if you don't snap or kink the center core.

This post has been edited by mzil: Nov 25 2012, 23:36
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Nov 26 2012, 02:53
Post #13





Group: Members
Posts: 4300
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (greynol @ Nov 25 2012, 12:27) *
QUOTE (Speedskater @ Nov 25 2012, 08:41) *
I have references for [b] & [c]

I'd be interested in a reference for [d].


Just to review:

d] RG-6QS (Quad Shield) is designed for the very high frequencies of cable TV so it's not the best choice for digital or analog audio. Ordinary RG-6 is a better choice.

I think that what is being referred to is the fact that most quad-shield RG-6 has what is known as a copperweld center conductor. This means that the wire is copper over steel (for strength)

This is, however not a 100% generality as I have seen quadshield RG6 that has a solid copper center conductor/

If you are putting a cable TV signal down the line, the lowest frequency is usually 50 MHz and skin effect is so strong that the fact that the center of the cable is steel which has only a fraction of the conductivity of copper, isn't an issue. Only the center is steel, and very little of the signal flows there. The signal flows mostly on the outside which is copper.

There are even more extreme cases of this sort of design of the larger cable TV cable (used for trunk lines) where the center conductor is an aluminum tube with silver plating. Again skin effect is so strong that the fact there is no conductor at all and just a hollow tube doesn't matter.
Go to the top of the page
+Quote Post
Soap
post Nov 26 2012, 03:00
Post #14





Group: Members
Posts: 1017
Joined: 19-November 06
Member No.: 37767



QUOTE (Arnold B. Krueger @ Nov 25 2012, 20:53) *
I think that what is being referred to is the fact that most quad-shield RG-6 has what is known as a copperweld center conductor. This means that the wire is copper over steel (for strength)


This is done for economic reasons only. I have in front of me copperweld RG6 and RG6QS. I also have solid copper RG6 and RG6QS.

So if you mean "most inexpensive cable" instead of "most" I'll agree, but I still want to see any evidence than the preponderance of RG6QS being copper-over-steel is higher than that of RG6.

This post has been edited by Soap: Nov 26 2012, 03:01


--------------------
Creature of habit.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Nov 26 2012, 03:01
Post #15





Group: Members
Posts: 4300
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (washu @ Nov 25 2012, 13:03) *
I've found SPDIF rather resilient when run over cables that are not to spec. I've got a 30' run that is made of a standard 1/8" phono jack extension cable (because it's what I had) with adapters on either end. The far end has a splitter to feed it into an old set of cambridge soundworks 4.1 speakers that needs separate inputs for the front and back pairs. A very hackish solution and definitely not in spec, but it works fine.


Science and practical experience backs you up.

First off, while SPDIF cables carry digital signals, the digital signals are not sharp edged square waves because consumer SPDIF outputs are generally fed though a low pass filter to minimize EMI so that the equipment can pass FCC Part 15 rules. Typical bandpass runs from 8 to about 20 Mhz.

Secondly, the impedance of a transmission line has negligible effect until the transmission line is 1/8 wave or longer at the highest frequency being passed. The wavelength of 20 MHz is about 15 meters or about 50 feet. The effect of any mismatch depends on how bad the mismatch is. A 2:1 mismatch is considered to be not that bad.

SP/DIF is generally very resilient, with outputs that put out a lot more than minimal voltages and inputs that require a lot less than specified voltages to work well. For example I'm driving two S{/DIF inputs in parallel with a simple Y cable in my AV system. Not pretty, but it gets the job done.
Go to the top of the page
+Quote Post
washu
post Nov 26 2012, 03:28
Post #16





Group: Members
Posts: 136
Joined: 16-February 03
From: Ottawa
Member No.: 5032



QUOTE (greynol @ Nov 25 2012, 13:08) *
Are you sure your speakers aren't being fed by a TTL signal?

I'm not sure what you mean by a "TTL signal" in this context, but they are definitely being fed by SPDIF.

The signal path is as follows:
HTPC Toslink out -> standard Toslink cable -> Marantz 6200 Toslink In -> Marantz 6200 SPDIF out -> long non-spec SPDIF connection and adapters -> Cambridge Soundworks speakers.

This allows the same music to be played on the speakers connected to the Marantz 6200 and the Cambridge Soundworks simultaneously. This only works for PCM signals, not AC3 or DTS, but that is a limitation of the Cambridge Soundworks speakers. The Marantz 6200 does not output anything on the analog outs when fed a Toslink or SPDIF signal.

I actually have a third and forth sets of speakers connected to another receiver that is fed from the Toslink out of the Marantz. That is over a 60 foot optical run which also works fine. A rather hackish full house sound system, but works great.
Go to the top of the page
+Quote Post
bennetng
post Nov 26 2012, 03:51
Post #17





Group: Members
Posts: 233
Joined: 22-December 05
Member No.: 26587



QUOTE
RG6 and RG59 look the same in photos, but RG6 is slightly larger and stiffer, which makes it harder to work with in a tight stereo cabinet. If the center core wire diameter is around 1.02 mm, then you have RG6, if it is .64 mm then you have RG59. Both should work well but be careful not to bend them too tightly; in best practice never more than 6 inch diameter circles should ever be attempted. Many people understand the danger of snapping the inner core, but what a lot of people don't realize is you also potentially can compromise the performance of the wire if it is bent more tightly than this, even if you don't snap or kink the center core.


Thanks. Because I only got replies like RGxx yesterday I only searched RGxx but today when I searched the text printed on my cable (5c2v) I got many results that put RG6 and 5c2v together. I wonder if they are the same?
Go to the top of the page
+Quote Post
mzil
post Nov 26 2012, 05:40
Post #18





Group: Members
Posts: 710
Joined: 5-August 07
Member No.: 45913



Your cable is marked with a JIS (Japanese Industrial Standard) designation , but yes, it is effectively the same as RG6.

This post has been edited by mzil: Nov 26 2012, 05:44
Go to the top of the page
+Quote Post
bennetng
post Nov 26 2012, 08:56
Post #19





Group: Members
Posts: 233
Joined: 22-December 05
Member No.: 26587



Thanks mzil!
Go to the top of the page
+Quote Post
2Bdecided
post Nov 26 2012, 13:52
Post #20


ReplayGain developer


Group: Developer
Posts: 5280
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



QUOTE (mzil @ Nov 25 2012, 21:49) *
Beware that, in some industries, no one really cares about the difference.

Some connectors will cross-connect, and others won't. There's quite a debate about it...
http://www.diyaudio.com/forums/parts/18200...-connector.html
http://www.vintage-radio.net/forum/showthread.php?t=47357
I have connectors that defy every piece of information in those threads, and others that match what's said, so... wink.gif

A general point is that, up to 10MHz, it doesn't really matter. On short runs, it doesn't really matter.


RCA/phono connectors were never designed to work at 6MHz, and often exactly the same cable is used for audio and video connected to phono connectors. On short runs, it's usually good enough for SD analogue composite video, and more than good enough for SPDIF digital audio. Cables designed to work with analogue HD video (or shall I say, cables that work really well with analogue HD video) are probably some of the best consumer cables out there. If they make great HD video cables, they'll make exceptional SPDIF cables. Though they normally come in triplets!

(I'm not claiming any audible difference from using better cables for SPDIF. I doubt there any detectable differences even in the SPDIF raw signal over short runs, never mind the audio that's decoded from it. On longer runs, you might reach a point where differences became detectable.)

Cheers,
David.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Nov 26 2012, 14:31
Post #21





Group: Members
Posts: 4300
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (Soap @ Nov 25 2012, 21:00) *
QUOTE (Arnold B. Krueger @ Nov 25 2012, 20:53) *
I think that what is being referred to is the fact that most quad-shield RG-6 has what is known as a copperweld center conductor. This means that the wire is copper over steel (for strength)


This is done for economic reasons only.


I don't know about that. The copperweld product I see has a lot more tensile strength, which might mean something outdoors and when strung from pole to house, etc. In general the outdoor cable TV system runs around here seem to be bare unsupported coax.

There is a standard RG-6 product with a built in steel support wire, but I don't see it used very often.

QUOTE
I have in front of me copperweld RG6 and RG6QS. I also have solid copper RG6 and RG6QS.

So if you mean "most inexpensive cable" instead of "most" I'll agree, but I still want to see any evidence than the preponderance of RG6QS being copper-over-steel is higher than that of RG6.


IME the cable TV industry are aggressive users of copperweld coax, these days mostly RG6. It seems like their consumption of coax has to be right up there if not more than everybody else put together. This might be especially true given their field tech's propensity for replacing all of the outdoor coax in customer systems every time there is a cough or a sneeze. ;-) My house has been rewired that way so many times I've lost count!
Go to the top of the page
+Quote Post
Soap
post Nov 26 2012, 14:37
Post #22





Group: Members
Posts: 1017
Joined: 19-November 06
Member No.: 37767



FWIW the fraction of my interior wiring done by the cable company was RG-6 QS, solid copper, and really nice shielding - 80+% braid. Quite expensive stuff. My pole drop at my current house and at my prior were both done with an integrated support wire - not that there is anything surprising about that seeing as it was likely the same Time Warner install sub.


--------------------
Creature of habit.
Go to the top of the page
+Quote Post
Dynamic
post Nov 26 2012, 20:39
Post #23





Group: Members
Posts: 832
Joined: 17-September 06
Member No.: 35307



You should get away with a lot with S/PDIF because a lot of the design criteria really help. The Bipolar Manchester Encoding of ones and zeroes makes thresholding at the mid-point easy and clock recovery very robust. The limited bandwidth lets you low-pass filter to remove nearly all the ringing seen in those avtechpulse diagrams if you should mismatch. Any attempt to match impedance, how ever poor, is likely to be enough. And the 0.2V minimum voltage swing at the input should be well clear of the range where Bit Error Ratio becomes even measurable.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 27th November 2014 - 08:07