Line level audio over CAT5

[mpkayeuk]

Hi,

Does anyone know of DIY project for sending line level analogue audio over CAT5? I tried last night by simply soldering some RCA plugs onto two pairs (for stereo) from a CAT5 patch lead. The output was OK but there was a noticable buzzing induced somewhere along the CAT5 cable run.

I know commercial baluns exist for this purpose and they seem very simple (I haven't actually purchased one but they look physically very small) but I would imagine there is some additional noise rejection componentry in there somewhere.

What can I do to make these myself (the baluns are pretty costly for what they are)? Any help greatly appreciated.

Thanks,

M

[resolution]

The baluns simply balance the audio.
There are many DIY solutions available online.
You need to convert to balanced audio at the send and then convert to unbalanced at the receiver.
You can do this with transformers (balun) or with active circuits based around op-amps. Obviously the balun does not require any power.

res

[paulb]

Have a look at this, if you haven't seen it already:
http://sound.westhost.com/project51.htm

Mercator on this forum is busy attempting to do this as well. He's had what I think might be some oscillation problems.

I'm using a DRV134 (from TI) to drive my audio cable throughout my house. You don't need a transformer (balun), and they may cause more trouble than they're worth anyway.

[dhaen]

Hi Paulb,

Balanced audio over cat5 is excellent, but you must accurately balance and unbalance to get the benefits.
I agree with the previous poster about the merits of the DRV134 (SSM2142).
You can get receiver chips to match SSM2141 or SSM2143.

Send me PM if you are interested in PCB's.

Cheers,

[AndrewT]

Hi,
does the question and the replies mean that the ONLY way to distribute analogue over CAT5 is by balanced techniques?

Will psuedo balanced at the send end suffice?

Can multi channels share the other pairs and still preserve crosstalk performance?

Can a non analogue signal use the spare pairs without ruining the analogue performance in the same cable or in an adjacent cable?

[paulb]

No, not the only way. The idea is to eliminate noise pickup from the cable. A balanced receiver will suppress most of the noise. However, each side of the balanced transmission needs to have the same source impedance for best noise rejection.
Did you read the article I recommended?
If you use the other pairs for other channels, you will get some crosstalk (although the balanced receiver should take some of it out). You should be able to run DC along the other pairs without too much degradation to the signal.

[mercator]

As paulb noted, I recently ran a couple of runs of cat5 from my server in the basement up to my living room - about 15m all told. I had initially tried using a pair for each channel as mpkayeuk did and had poor results - low signal and noisy. Also, my amp got pretty hot - possibly due to oscillation(?).

So, I built a differential line driver unit based on drv134 chips (using digi01's excellent boards). At present I don't have differential receivers (that is a feature of the preamp currently under construction ) so I am just using one side of the differential pair. I can report that it seems to work well and sounds good, however, when I first turned it on, after about an hour of listening, two of the four output devices on the left channel of my amp (P3A) went pfffft

I am not certain there is a direct correlation between using the cat5 and my amp failure. I hooked up a cheapo sony amp for tunes in the meantime and it has been working without incident for about 40 hours now.

This is still a work in progress but I think the cat5 is not going to be a problem.

/Dave

[AndrewT]

Hi Mercator,
you seem to be saying that the balanced line driver feeding into balanced cable but then into unbalanced phono on the power amp has cured the noise & signal level problems encountered using unscreened twisted pair in the CAT5. Or have you fed half the balanced signal into the twisted pair with the unbalanced phono at the recieving end?

Surely the failure of the P3 output stage can only be co-incidence, since the input from the CAT5 cannot cause instability in the power amp??? Is there an RF filter on the input?

[mercator]

Hi Andrew,
I am feeding half the balanced signal into the twisted pair with unbalanced phono jacks at the other end, and yes the noise and signal level problems are (more or less) gone. You might conclude from this that the solution was not the balanced connections so much as having more robust line drivers provided by the drv134s.

Regarding the amp failure, I agree it is probably coincidence. Hopefully the amp will be running again this week and I will check the signals on the scope before I hook it up again. The oscillation could have been coming from the sound card.

As for whether the amp has RF filters on the input, I don't know the answer to that, it is built according to Rode's specs, on a stock board, the schematic is here.

Cheers,
/Dave

[AndrewT]

Hi,
R1 & C2 form the RF input filter about 0.5uS which is OK.