CAT 5 analogue snakes

[ajay_m]

I see these are becoming increasingly popular e.g https://www.thomann.de/gb/the_sssnake_cat_snake_3mc.htm

I was curious; as a decidedly low tech option four of these plus four of the male XLR variant lets you route 16 channels of audio over four runs of cheap cat5 cable, with the huge advantage that if somebody chews up one of your cables while rigging, or whatever, you can just whip out a reel of cat5, cut to length and quickly crimp connectors on it and you're good to go. Much cheaper than Dante and a stagebox. But does this really work, at least for fairly short runs, at mic level? Just curious.

[The Elf]

I have a couple of these as stage backups. I've never used them in anger, but they've worked fine when tested.

[Aled Hughes]

Yup, I have the Thomann ones and they work as advertised. Only “issue” I’ve encountered was crosstalk when sending an unbalanced stereo signal down a single channel, but that’s not a fault with the product as such, and they don’t claim it should be used that way anyway - strictly speaking it’s designed to carry four balanced mono signals. It was still usable doing what I did though.

Currently waiting for more to arrive, to eventually route signals to ceiling speakers.

[Hugh Robjohns]

...route 16 channels of audio over four runs of cheap cat5 cable, with the huge advantage that if somebody chews up one of your cables while rigging, or whatever, you can just whip out a reel of cat5, cut to length and quickly crimp connectors on it and you're good to go

Kind of...

As a minimum this systems needs screened cable with screened connectors — not the cheapo cat5 utp and unscreened RJ45s.

Ideally you need cable with individually screened pairs — especially if working with mic signals — which effectively means cat6 f/ftp with Ethercon plugs.

But yes, when setup right, it works well and is cost effective.

[ef37a]

It would be interesting I think to know Hugh if individual screening of CAT 4pairs does in fact give a significant improvement in crosstalk given that such cable is more expensive and stiffer?

At the impedance that mics work at I would think most of the coupling is magnetic rather than electrostatic, at the network frequencies the shield will of course be more effective.

Dave.

[Hugh Robjohns]

It would be interesting I think to know Hugh if individual screening of CAT 4pairs does in fact give a significant improvement in crosstalk

I don't have any suitable ftp and f/ftp cable here to make comparison tests. However, makers of analogue multipair çables wouldn't add individual pair screens if they weren't necessary.

That said, I have tested individually screened two-pair versus star-quad, and found there was a small but measurable difference in inter-pair crosstalk of a few dB (being worse for the star-quad)

In practice, the slightly degraded crosstalk isn't an issue when working with similar level and related signals (such as a stereo pair)... but can be with unrelated signals at dissimilar levels (or if working with unbalanced signals, including headphone feeds).

Given the OP is talking about live sound rigs, where signals passing up and down the cables can vary in level and format enormously, I'd definitely want individually screened pairs.

[ef37a]

"However, makers of analogue multipair çables wouldn't add individual pair screens if they weren't necessary"
Oh I am not disputing that Hugh, as I mentioned in my original question. I just wonder how effective capacitive screening is at low impedance and audio frequencies.
I shall see if I can find anything online and yes of course, much depends upon the mix of signals. It would not be a good idea to send mic and +4dB line signals down the same 4 pair.

Dave.

[ajay_m]

I meant shielded cat6 cable by the way, sorry not cat5.

[ef37a]

I meant shielded cat6 cable by the way, sorry not cat5.

There seems to be two types of CAT 6 shielded cable. One just has an overall foil shield with a drain wire and another has individually screened pairs with a braided overall shield. My question is, is it worth the extra for individual shielding? It might seem blindingly obviously better but I do wonder at audio frequencies?

Of course you can also get CAT 5/5e overall shielded cable and that is going to be just as good for audio purposes. I have a lot of it around my bungalow!

Note too that there are two typed of conductors. "Backbone" fixed site cable uses solid conductors whereas "patch" cable is stranded for extra flexibility. You need the correct RJ45 plug for each type. Patch cable is more expensive but solid core is in fact very robust and I have never had a problem with it.

Dave.

[ef37a]

https://www.youtube.com/watch?v=gGgPDEz5sNU

Make of that what you will but I cannot see very much wrong with his methodology?

Dave.

[Hugh Robjohns]

Really? With the mics feeding back on the monitor speaker? Nothing wrong with the methodology?

But as I've said previously, and demonstrated using star-quad pairs HERE, crosstalk is typically -100dB or better and irrelevant... under ideal circumstances.

However, one of the methods of managing crosstalk in Ethernet cable is to use symmetrical balanced signals. This is because the opposite polarity signals on the tightly twisted wires in each pair largely cancel out the magnetic and electrostatic fields that would be created by a single wire, thus hugely reducing induction into adjacent wires, minimising crosstalk.

And in Ethernet connections all sources generate symmetrically balanced outputs because the standard demands it.

In audio, this is NOT guaranteed to be the case. Most people think of balanced audio as being symmetrical signals in opposite polaritieson each wire, but very often it's not.

So called 'impedance-balanced' outputs are commonplace on budget mixers and a lot of other gear besides — including many modern transformerless capacitor microphones from very respected big name manufacturers.

Sending single-sided (impedance-balanced) signals down an unscreened pair will generate a significant magnetic field which is likely to induce a much higher level of crosstalk into adjacent unscreened pairs.

So.... under ideal conditions where the source equipment has symmetrical balanced outputs, and signal levels are all similar, I'm know unscreened CAT cable works acceptably.

However, in challenging, unpredictable circumstances in live-sound, with unknown source, hugely variable signal levels, I personally wouldn't risk my reputation on it, and would always use individually screened pairs if running analogue over CAT cable.

Running Ethernet data for a Dante system etc, would not normally require individually screened pairs.

[ajay_m]

Someone reviewing the Thomann units said phantom power didn't work but I can't for the life of me see how that could be the case, it'd take quite a high impedance wiring connection to drop 48V at a few milliamps too low at the other end

[Sam Spoons]

It would only take for the screen connection to be OC at one end to defeat phantom power, the signal would still get through from devices that didn't need spook juice.

[Aled Hughes]

Someone reviewing the Thomann units said phantom power didn't work but I can't for the life of me see how that could be the case, it'd take quite a high impedance wiring connection to drop 48V at a few milliamps too low at the other end

I think phantom power relies on using shielded cable. If you use unshielded, I think it will still carry audio but phantom won’t work.

And just in case it was my fault for bringing yp the crosstalk- I’ll emphasise again that I was not using the device as intended!

[Sam Spoons]

Yes, unshielded cable could not transmit phantom power.

[Hugh Robjohns]

Someone reviewing the Thomann units said phantom power didn't work but I can't for the life of me see how that could be the case, it'd take quite a high impedance wiring connection to drop 48V at a few milliamps too low at the other end

Nothing to do with impedance. The muppet was using cheapo UTP (unscreened) Cat5e cable.

No screen means no return path. Hence no phantom volts at the far end.

[Hugh Robjohns]

There seems to be two types of CAT 6 shielded cable.

There are several variations, all with different code letters, although different manufacturers sometimes use different codes!

All ethernet cables contain four twisted pairs of wires, each pair carrying one differentially encoded signal. So all cable codes end in TP for twisted pair.

The simplest cable type just wraps those four pairs in an outer sheath, so it is unshielded and is called UTP

The next level up is to have an overall shield just under the outer sheath. This can be foil (F), or a braided screen (S), hence STP or FTP

Another option encases each pair in its own shield (always foil), omitting the overall sheath shield, coded as U/FTP — meaning unshielded overall, but foil shielded pairs.

Then there are full monty versions with either braided or foil overall screen as well as foiled pairs, coded as S/FTP or F/FTP or sometimes SSTP (Shielded Screened Twisted pair).

...other codes may describe the same arrangements...

[ajay_m]

Surprisingly cheap; a 300 metre reel of outdoor-use S/FTP Cat6a cable is £200 doing a quick google. (probably cheaper vendors, that was just the first one I found that looked reasonably reputable).
And you'd suppose the 8 adaptor thingies you purchase for a 16 channel setup will last a long time because they get unplugged and put safely in a bag at the end of each gig. I always love it when a piece of technology gets appropriated for a purpose it was never remotely intended for - I mean, analogue over Ethernet. I had never ever thought of that! All those fancy digital protocols like Dante and MADI and Ravenna all that (of course they support much higher audio counts per cable) and then someone just comes along and does this - the ingenuity of the human mind never fails to amaze me!

[ef37a]

"Sending single-sided (impedance-balanced) signals down an unscreened pair will generate a significant magnetic field which is likely to induce a much higher level of crosstalk into adjacent unscreened pairs."

Yes I originally asked if we needed individually screened pairs for low crosstalk with audio and that video seemed to suggest not. In any case I cannot see a thin foil having much if any effect on electro MAGNETIC radiation? Mumetal might.

But until someone does the tests we shall not know. Meantime I understand your reservations Hugh, especially with impedance balanced sources (although many of these CAT audio systems use transformers)

BTW just in case anyone else was initially thrown by "Muppet using UTP cable"? The guy in the YT link I gave was definitely using overall shielded cable.

Dave.

[Hugh Robjohns]

But until someone does the tests we shall not know.

I don't have any accessible cat6 cable types to hand for testing. However, I do have lengths of two channel multi-pair and star-quad wired for two-channel use which should give some illustrative figures. I'll try and do some measurements next week.

... (although many of these CAT audio systems use transformers)

The balun types designed for unbalanced audio and video do. However, the fpur channel XLR systems discussed here are simple wired adaptors without transformers.

BTW just in case anyone else was initially thrown by "Muppet using UTP cable"? The guy in the YT link I gave was definitely using overall shielded cable.

I was referring to the chap who claimed the system didn't pass phantom.

[AlecSp]

There seems to be two types of CAT 6 shielded cable. One just has an overall foil shield with a drain wire and another has individually screened pairs with a braided overall shield. My question is, is it worth the extra for individual shielding? It might seem blindingly obviously better but I do wonder at audio frequencies?

Cable with individually shielded pairs will cost more, and handle less well than regular SFTP cable, depending on how much that matters to you.

I always love it when a piece of technology gets appropriated for a purpose it was never remotely intended for - I mean, analogue over Ethernet.

Of course, it's not "analogue over Ethernet", it's analogue over Cat 6 (or otherwise) cable - too many people are confused about protocols.

[ef37a]

Yes Alec, plenty of scope for confusion but "that's the name of the game" People still need to be reminded sometimes on forums that MIDI is not audio!

As for analogue audio over what are essentially telephone pairs? The BBC were sending high quality (for the day) audio over GPO lines for decades, though I understand that they had some pretty fancy-ass EQ systems both ends? Don't know if crosstalk was a problem or if they could use dedicated single lines? I remember building "Telco" panels which carried a thumb thick 48pair cable.

Dave.

[BigRedX]

The big problem I see with this system is I have yet to discover a Cat5 or Cat6 cable that I would consider gig-proof. Used to own a Bass Pod plus Floorboard which were joined with Cat5 cables. Standard installation Cat5 lasted less than a month. Two very expensive Van Damme coilable Cat5 leads last about a year each. This was one of the reasons why when I upgraded my effects unit to a Helix I went with the floor version so I wouldn't have to rely on computer grade components any more.

If I was using a system like this I would want plenty of spare leads or be absolutely certain of my crimping skills under pressure in a dark environment.

[Hugh Robjohns]

I far as I can remember, the fancy-ass EQ was only at the destination . And the lines were called 'Music Lines' which were essentially telephone lines selected for good quality.

[ef37a]

The big problem I see with this system is I have yet to discover a Cat5 or Cat6 cable that I would consider gig-proof. Used to own a Bass Pod plus Floorboard which were joined with Cat5 cables. Standard installation Cat5 lasted less than a month. Two very expensive Van Damme coilable Cat5 leads last about a year each. This was one of the reasons why when I upgraded my effects unit to a Helix I went with the floor version so I wouldn't have to rely on computer grade components any more.

If I was using a system like this I would want plenty of spare leads or be absolutely certain of my crimping skills under pressure in a dark environment.

I think the great advantage of CAT cabling is that it is so cheap that you can afford to carry a great many spares. I would also aver that RJ45-RJ45 cables are far more robust than MIDI DINs? And if "crimping in the dark" fazes you try soldering a DIN plug? Even in daylight! Also, make sure you fit "no snag" booted plugs.

Dave.