Audio over ethernet

M

mikeand

New member
Hi all,

I am just looking into a way to send 24-32 channels of line audio over Ethernet from a mixer to a set of patchbays and or ADAT recorders. I have been browsing the gear manufacturers (MOTU, Digigrid, Focusrite, Steinberg etc) for devices to do this. But what I could really do with is some advice from someone who has tried any of the devices out. I only need to send the audio about 15-20 meters and it strikes me that Ethernet is the way to go rather than making a 32ch 15mtr loom.

Any wisdom much appreciated,
Cheers,
Mike
 
Probably Dante is the thing to look at now as it's available on many devices. If your mixer is one that has the capability then it's perhaps worth looking? The snag is that Patchbays and ADAT are just not modern enough to do this without connecting to other kit that probably you don't want? if your mixer has it's own ethernet connection like the Behringers then converting the other end is a possibility - it's not network of course, just uses the same cables and connectors.
 
Whatever you use, it isn't going to be cheap if you don't have anything with ethernet already built in. Since one of the destinations appears to be ADAT then maybe you could use some of the Behringer ADA8200 convertors on one end feeding the ADAT machines via optical cable (although strictly speaking the limit on ADAT optical is 5m). Otherwise you need to convert the analogue signals to digital which isn't cheap. Ferrofish have 16 in and 16 out convertors but they are £1200 if you want MADI or over £2000 if you want Dante as well.
 
HI guys, thanks for those suggestions, I will start to look at all of them, not ones I have looked at so far, so very helpful.
Cheers,
Mike
 
bouldersoundguy said:
You can send four channels of balanced analog audio down one Cat5/6 cable using adapters from various manufacturers.
Click to expand...
You can indeed and if each output is balanced and of decently low source impedance Mike should not need any additional gear. Even if a digital Ethernet system were used you would still have to go to analogue cables to the patch bay.

The various bit of kit could feed into a box and exit on 4 Tascam wired D subs each containing 2 shielded CAT5e patch cables. At the patch bay it will depend on how that is terminated. Best option IMHO is a solder job as that means no more connectors.

You will need to draw up loads of diagrams to plan the connections...Sort of thing I LOVED doing!

Dave.
 
Thanks guys for all these hints,
I will have a look over that video. Thanks TAE
Do you have any recommendations for adapters sending four channels down a cat5/6 cable.
ecc83 - a fan of diagramming myself - makes we all warm inside -sorry that sounded less weird in my head :P !!
 
Isn't this going to introduce latency? I suppose if only mixing/playing back, then it matters not. But if expecting to be doing anything in real time while tracking I would assume there's a penalty adding converters into the chain.
 
If you can get your hands on a single ADA8200, you could connect a 20m optical cable in a loopback arrangement, to test whether optical ADAT can go that far.
The experiment would also reveal the latency.
Even with cheap behringer kit, 4 ADA8200s at each end, would get pricey.

Experimenting on sending 4 channels of balanced audio down a cat5 cable, would be cheap to try. I worry about it being unscreened.
I have heard of people shoving analogue video down cat5 cables.
This would be a cheap option , if it worked.

The most effective option would be acquiring the right box for each end, and sending the digital data over ethernet, which for sure is fast enough.
But this woud be big money.
 
Slouching Raymond said:
Experimenting on sending 4 channels of balanced audio down a cat5 cable, would be cheap to try. I worry about it being unscreened.
I have heard of people shoving analogue video down cat5 cables.
This would be a cheap option , if it worked..
Click to expand...

You could use STP (shielded twisted pair). That said, I've been using UTP (unshielded) for 8-10 meter runs of balanced line level with no problems.
 
Slouching Raymond said:
Hang on a minue. You don't need to make a loom. Just buy a ready made snake, or a digital snake.
Click to expand...
Yes, fine Ray but we don't know what all of Mike's source/sink connectors are, mix of TRS, XLR, GKW? I have had a quick look for 20m TRS-TRS and an 8 way is a nifty and he needs 4 of those.

Going Ethernet means an A/D send to the patch bay. D/A AT the bay then A/D from the bay and finally D/A back at the inputs. Does such a set of boxes exist and that is four lots of latency as has been mentioned. Not quite the same thing as an "E" stage box to a FOH mixer which might be digital in anyway? Is there such a thing as a digital "selector-patch bay"? I know there are digitally controlled analogue patch systems.

BTW Mike, whilst I do enjoy the planning of such cabling systems I meant I really enjoyed the actual building of the stuff!
As I am sure you realize, a very effective cable labeling system will be needed.

Dave.
 
bouldersoundguy said:
You could use STP (shielded twisted pair). That said, I've been using UTP (unshielded) for 8-10 meter runs of balanced line level with no problems.
Click to expand...
Beat me to it BSG. You would think only 'patch' FTP would be suitable but if, as I strongly suggest, the cables go to boxes, either in them or on D subs, it is quite in order to use solid core pair "backbone" cable. The stuff is easily flexible enough and I don't think I have ever had a wire break. Solid pair is much cheaper than flexible.

N.B. CAT5/5e has one overall screen and a drain wire. PofP to work with. CAT6 has individual screened pairs and it much stiffer and a bit of a PITA to strip. The better digital performance of CAT6 is of course not needed for audio.

Dave.
 
Yes Rob but they are based on transformers and whilst fine
for a one-off "PA" situation I would not want to use them in a high quality studio setup? I cannot see how they can justify the (fabulous!) specification since a transformer's response and distortion will be very dependent on the source and load resistances. THD might be very low at 1kHz but at 40Hz it could be 1% at +4dBu! Then there is induced hum.

Mike is also going to need a hell of a lot of them, to'ing and fro'ing a patch bay...


Dave.
 
You must log in or register to reply here.
Back
Top