Audio Cable used for CAT 5 /6?

Help Support GroupDIY:

john12ax7

Well-known member
Joined
Oct 14, 2010
Messages
1,930
Location
California, US
The network connection worked that way but only at 10Mbps

This is the most likely outcome. The higher the frequency (bit rate) the more the details matter, and the more you need the proper cable. Sub-optimal solutions can be perfectly fine at lower frequencies.
 

berret

New member
Joined
Aug 26, 2009
Messages
3
A couple of years ago i did a live broadcast for Dutch Classical Radio, Radio4, from the small concerthall in the Frits Philips Muziekgebouw ,Eindhoven the Netherlands. Just before the start of the concert the stagehands moved the grand piano and it went over the multicable. Suddenly i missed the connections from my main stereo mics and the pianomics. And in a couple of minutes i routed the signals through the existing cat5 STP connections from the stage to the controlroom. And i could not believe my ears: the sound was even better compared to the expensive multi. Since that time i am convinced from the benefits of a good STP cat5 cable.
Just my 2 cents.
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
119
Location
Ventura, CA
Something missing from this thread (except for a mention in passing of 100 Ω, 110 Ω, etc.) is "characteristic impedance" of cables, which is what those numbers are. Briefly, characteristic impedance becomes important when the physical length of a cable becomes a significant fraction of the electrical wavelength of the signal. Characteristic impedance of a cable is a function of the conductor diameter, conductor spacing, and the dielectric between them. The ubiquitous 600 Ω originated in the characteristic impedance of two #10 AWG conductors, spaced about a foot apart in open air on telegraph wires. The guys at Bell were smart enough to know they needed to terminate each end of the pairs in 600 Ω to avoid reflections that would cause severe aberrations in frequency response. The wavelength at 20 kHz (in typical cable where signals travel at about 90% of light speed) is about 7 miles, so an audio cable longer than about 3,500 feet needs to be "terminated" in its characteristic impedance. And, for typical mic cables (whose twisted pair conductors are of heavier gauge than CAT cables), characteristic impedance is about 60 Ω, BTW. As frequencies get higher, wavelengths obviously get shorter, so 100 Mb/s data cables longer than a few inches must be terminated to avoid reflections. Such reflections from the mismatched end of a cable will cause standing waves (the same kind your "ham radio" buddies talk about) and cancellations that will cause extreme errors in data signals. Not only that, the insulation used in CAT cables has very low dissipation factor (loss) at MHz frequencies. So signal losses in ordinary audio cables will be huge. There's a lot more to it than just twisting and CMRR - twisting does little about capacitive crosstalk unless the line driver and receiver have very good impedance balance. So unless a cable run is very short, and the data rate very low, ordinary twisted pair designed for audio frequencies will be nearly unusable as data cable. Using data pairs for audio is trivial, but going the other way is just asking for trouble. Wireless solutions might be the best work-around. Just my 2-cents worth!
 

john12ax7

Well-known member
Joined
Oct 14, 2010
Messages
1,930
Location
California, US
The wavelength at 20 kHz (in typical cable where signals travel at about 90% of light speed) is about 7 miles

Curious where you are getting the 90% number from? The velocity would depend on the dielectric used. Typical run of the mill cables tend to be in the 65-70% range. I looked up Mogami 2549 and it uses polyethylene, so would expect around the usual 65%.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,994
Location
Marcelland
Something missing from this thread (except for a mention in passing of 100 Ω, 110 Ω, etc.) is "characteristic impedance" of cables, which is what those numbers are.
It looks like you haven't read my #9 post.
Your claim that "for typical mic cables characteristic impedance is about 60 Ω" is not in accordance with the AES conclusions, that recommended 110 ohms for AES/EBU digital connections, because it allowed the use of standard mic cables. It is not in accordance either with many cable brochures.
Indeed, some cables, like the Canare Starquad, have twice the capacitance, which results in about halving the characteristic impedance, but they cannot be considered "standard" mic cables.
Of course, there are other parameters that may make audio cables unsuitable for data use, but in the absence of precise requirements, we're just speculating.
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
119
Location
Ventura, CA
The 90% number would apply to foam dielectrics, which is presumably what a mile-long run would use, just for the sake of low capacitance. My use of the word "typical" was inappropriate. The 60 Ω characteristic impedance I quoted was actually the optimum driving source impedance for common STP audio cables like Belden 8451, not its characteristic impedance. From the same AES paper, "... Belden 8451 cable which is a single twisted pair of 22 gage stranded wire, with foil shield, drain wire and an overall jacket of 3.43mm (0.135 inch). A rough calculation of the pair's characteristic impedance based on wire size, dielectric constant, and insulation thickness (pair spacing) without regard to the shield shows it in the neighborhood of 175 ohms." (emphasis mine, see Voltage Audio Distribution) I remembered that Richard had done a paper about this but didn't get my numbers right. In any case, you won't get much high-speed data through audio twisted-pairs that aren't made with a specified characteristic impedance. For AES-3, there are a number of STP cables with 110 Ω twisted pairs just for the purpose. Again, they make good mic cables but not vice-versa, only shorter runs will tolerate the mismatch.
 

kaltavilla

Member
Joined
Dec 18, 2017
Messages
6
Really? What is AES/EBU? Isn't it data?

I would use it if I had no speaker cable and enough coat hangers.
When I was on tour in a very poor island, I was short on mic cables; I used standard 2-core electric cable, which was all that was available, and used it for the loudest mics.
Of course, I wouldn't do it for permanent use.
I understand, we all do what we have to to get the job done. Nothing to apologize for there.

Regarding AES/EBU cable, it is a low capacitance, twisted, shielded pair, whit a characteristic impedence that has been engineered to meet the need of the AES/EBU spec (110 ohms). That said, of course it works as audio cable. It works very well as audio cable because the capacitance between pairs is almost half that of your basic audio pair, so you end up with less High frequency (relative to the audio range, below 20k Hz) loss on longer runs. It does sound better than 'regular' audio cable.

But 'regular' audio also works as AES/EBU cable, just not as well, and not on long runs, because of the impedance mis-match.

The OP wants to run DATA, down AUDIO cable. Same deal as using audio cable for AES/EBU. Yeah it will work, but not well, and not on a run longer than maybe 10 ft., and certainly not at high speeds.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,994
Location
Marcelland
Regarding AES/EBU cable, it is a low capacitance, twisted, shielded pair, whit a characteristic impedence that has been engineered to meet the need of the AES/EBU spec (110 ohms).
Actually, it's the other way round. 110 ohms has been chosen by the AES because it's the typical characteristic impedance of audio cable.
Actually, I've checked the Belden catalogue and they specify most of their audio cables at 45 ohms. I don't know what in the construction of their cable it is the result of. I've dealt with many brands and types of cables, and most confirmed the 100-120 ohms ballpark. For comparison Mogami cables have a capacitance of about 80pF/m and a characteristic impedance that is about 105 ohms +/-10%. Canare cables are very similar.
That said, of course it works as audio cable. It works very well as audio cable because the capacitance between pairs is almost half that of your basic audio pair
so you end up with less High frequency (relative to the audio range, below 20k Hz) loss on longer runs. It does sound better than 'regular' audio cable.
100m (330ft) of audio cable put the -3dB point at 110+kHz with a 200 ohm source. Using AES/EBU cable would extend the BW to 190kHz. I can't hear the difference.
Is it a concern? Anybody that connects a mic through 100m of cable is a fool looking for trouble, or a desperate man.
The OP wants to run DATA, down AUDIO cable. Same deal as using audio cable for AES/EBU.
Still we don't know what kind of DATA he wants to transmit and over a distance of about 150ft.
Instead of claiming it can't work, I just say: "Since you have the cable, try it." What's the risk?
 
Last edited:

Whoops

Well-known member
Joined
May 9, 2008
Messages
5,688
Location
Portugal
From my searching, multichannel CAT 5 /6 cable is difficult to come by and very expensive, and I have ample audio cable.

I would use multiple individual cat5/6 cable rather than using audio multicore cable for that, anytime
 

Brian Roth

Well-known member
Joined
Aug 20, 2005
Messages
1,848
Location
Salina Kansas
Just my 2 cents after reading the OP's post in message 14 in this thread. More than a few years ago, a new studio build installed the (then new) HEAR cue system. That system used CAT5 cable, and I ASSume the data stream is a proprietary format. We "went by the book" and installed CAT5 cabling to all the sockets in the tracking room (Neutrik sockets on the wall plates).

Much later, I worked on another studio build with HEAR cue boxes and I specified CAT6 cable because it was readily available for a decent price.

Both systems worked perfectly....although the Hearback boxes are not all that robust in the long run.

For 150 foot runs, suck it up and follow the manufacturers specified CAT5/6 cabling design. We have no idea of the data format/speed down the cable since it's apparently proprietary. The cabling was NOT a problem....just the studio boxes are far from bullet-proof.

Bri
 

Whoops

Well-known member
Joined
May 9, 2008
Messages
5,688
Location
Portugal
What is your point here? Is it the superiority of CAT5/6 over audio cable or the possible savings made in using individual cables instead of multicore?

Hi Abbey, Both.
If the recommended cable for that protocol is CAT5/6 I would use whats correct and recommended for that specific protocol.
If multicore CAT5/6 is expensive, I would use individual cables as they are inexpensive.
 

Marcedona

Member
Joined
Sep 14, 2017
Messages
9
A side note.

When I was young (that phrase with Dion or Zazarashvili voice), I improvised a sound hose with industrial multipair cable.

The lines were twelve pairs of two live and one mesh each pair, like stereo, but I needed more; so I had the idea of using the live ones independently and doubling the channels.

I didn't notice any problems at home, when I tested the 24 channels one by one. - Already in the sound check I verified that when the voice raised the kick drum also raised, and when I tried to lower the guitar the keyboard also lowered, and so ... Advancing the show I verified it in a sadder way, while the musicians raised the volume of their stacks without any criteria.

I was young, I needed the money, I didn't know what I was doing.
Now, on the other hand, I'm not young haha. But generally using the things for what they are made for saves some trouble. Greetings!
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
20,969
Location
Hickory, MS
As usual a wealth of knowledge shared here. I wrote a dumbed down (for both me and my audience) version of this in my old "Audio Mythology" column, decades ago.

This is not very important at audio frequencies, but everything matters in the margin. To maintain integrity of managing digital signals characteristic impedance of cables and terminations do matter.

I recall running digital (midi) signals through audio snakes last century because it worked, while not ideal.

JR
 

HP200CD

Member
Joined
Feb 1, 2015
Messages
8
Location
Oyster Bay, NY
Really? What is AES/EBU? Isn't it data?

I would use it if I had no speaker cable and enough coat hangers.
When I was on tour in a very poor island, I was short on mic cables; I used standard 2-core electric cable, which was all that was available, and used it for the loudest mics.
Of course, I wouldn't do it for permanent use.
AES/EBU cable is not quite the same as balanced audio cable. AES cable is built to different specifications. AES cable can carry analog audio as well as (if not better than) balanced audio cable. On the two web pages below, note the cable capacitance (in picofarads per meter) and that the capacitance of the analog cable is about 4 times as much as the capacitance of the digital cable. The higher capacitance of the analog cable degrades the HF response of the digital signal to the point where it is unusable in a much shorter distance. Also note that the twist on the digital cable is almost twice as tight as on the analog cable.

Analog cable (StarQuad)

Digital cable (AES/EBU)

Also what MrCMRR said about reflections and standing waves in post #23 is very significant.
 
Last edited:

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,994
Location
Marcelland
AES/EBU cable is not quite the same as balanced audio cable. AES cable is built to different specifications. AES cable can carry analog audio as well as (if not better than) balanced audio cable.
Maybe you weren't there, but when the AES and the EBU teamed to define the AES3 standard, they defined it as usable on existing wiring, and based the standard on the fact that most of audio cables had a characteristic impedance of about 110 ohms.
For the same reason, Philips and Sony chose 75 ohms for S/PDIF, because most of the RCA cables have a characteristic impedance of about 75 ohms.
It's only later that cable manufacturers produced specific AES/EBU cable; which ensured that the characteristic impedance was more strictly controlled than for audio cable, where it's a non-specified parameter.
On the two web pages below, note the cable capacitance (in picofarads per meter) and that the capacitance of the analog cable is about 4 times as much as the capacitance of the digital cable. The higher capacitance of the analog cable degrades the HF response of the digital signal to the point where it is unusable in a much shorter distance. Also note that the twist on the digital cable is almost twice as tight as on the analog cable.

Analog cable (StarQuad)

Digital cable (AES/EBU)
You're comparing AES/EBU cable with Starquad, which is notorious for its high capacitance. Apples and oranges.
Indeed the twist on Starquad is higher because there are 4 wires instead of 2.
Also what MrCMRR said about reflections and standing waves in post #23 is very significant.
Reflections and standing waves are the result of inadequate matching of impedances. It is clear that using high capacitance audio cable results in a characteristic impedance that is lower than the recommended 110 ohms.
Please note that I'm not advocating using audio cable neither for AES/EBU nor data,; I'm just saying that the proof is in the pudding.
There are enough examples of using "wrong" cable that work satisfactorily. In the particular case of the OP, I just suggested that experimenting was the only way to know if it could work or not.
A system with "wrong" wiring will not work as well as one with the proper wiring, it will eventually cease to work, but knowing the limit without experimenting is very uncertain.
If the audio cable is of close enough impedance, the difference in performance will be related to the loss factor. The capacity of the receiving equipment to recover/process the signal is fundamental in the resulting performance. They're all supposed to be equal, but some more than others...
 

Gold

Well-known member
Joined
Jun 23, 2004
Messages
2,743
Location
Brooklyn
Maybe you weren't there, but when the AES and the EBU teamed to define the AES3 standard, they defined it as usable on existing wiring, and based the standard on the fact that most of audio cables had a characteristic impedance of about 110 ohms.
That's valid for 44/16 and maybe 44/24. Remember that early 24/96 systems used two cable runs? My ancient converters still support both specs. I haven't run the numbers but I'd assume 24/96 over a standard mic cable wouldn't work very well.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,994
Location
Marcelland
That's valid for 44/16 and maybe 44/24. Remember that early 24/96 systems used two cable runs?
I also remember atime where any AES/EBU connection was fragile, mainly because of the receivers. It has changed a lot.
I haven't run the numbers but I'd assume 24/96 over a standard mic cable wouldn't work very well.
Why don't you try instead of speculating?
 

Gold

Well-known member
Joined
Jun 23, 2004
Messages
2,743
Location
Brooklyn
I also remember atime where any AES/EBU connection was fragile, mainly because of the receivers. It has changed a lot.

i remember that the Weiss BW102 system was very particular about cabling. I can say with certainty a standard Mogami mic cable wouldn’t work with those even at 44.1K. If you had more than one XLR connector in line, even with AES cabling, was too much of an impedance bump for it to handle.
Why don't you try instead of speculating?
Both because I’m lazy and these things are often more about implementation than theory.
 

Latest posts

Top