Audio over twisted pair (CAT 5E) cable!?!

[bandcoach]

I remember reading something from a BBC engineer on-line last year about how they only used twisted pair (no shield) to wire up the audio side of their studios in the 50's and 60's. So I did a little research today and the current thinking seems to support the experiential advice offered by the BBC guy, as long as you use solid core Cat5E. The beauty of Cat5E is that you can have four channels per cable if necessary. It also mitigates ground loop problems that my be carried by an audio shield because there is no shield, but it does not eliminate ground loop problems completely. These exist because you have two or more power circuits that your gear is connected to that have separate earth points in the wiring system leading to voltage/current drift problems between the two or more nominal ground points.

So thinking about wiring up different parts of the studio using this now, particularly the balanced runs to and from gear. A single cable allows you to connect stereo send and returns on appropriate gear.

I'm also thinking about using terminated 24 point wiring frames at each end of the runs and using fully wired Cat5E cables to join them up. This means one frame supports up to 96 individual runs. At about $75 per 1000ft for Cat5E this makes a lot of financial sense as well. Running lines from the termination frames back to the gear would be done using regular shielded pair cabling (mic/signal cable) and connectors.

Yes, it's a custom fit-out, but certainly cheaper than doing the same thing using balanced audio cable throughout and with the same results.

thoughts/comments?

Shane

[bandcoach]

So, doing some investigating (again) last night and ended up at the following article

Balanced line driver and receiver taking single ended signals (guitar outs, most synth outs, etc) and converting them to balanced outputs and vice versa. Came out of following an old forum post from here - thank you again Guttermouth.


Additional info can be found here and here for the line driver and here and here for the line receiver.

These chips only require a power supply and two additional capacitors to achieve the results required.

Tie a set of these into driving and receiving signals over the Cat5E cabling, and everything stay nice and neat. This is twofold

• because every signal is able to make use Common mode rejection of stray noise (the noise is in anti-phase on the twisted pair lines, so is canceled out when the line receiver converts back to single ended signals,
• whilst traveling down the line, everything is running at 6db gain (double the signal strength), ensuring that any noise on the line is reduced even further (the line receiver drops the level by 6db) when converted back to a single ended signal by the line receiver

So, will spend some time designing a quad or octet printed circuit board of drivers and receivers for implementation with a pair of Cat5E RJ45 cables. Updates as things get done

Shane