4x XLR through Cat-Cable

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

juliusbusch

Member
Joined
Jan 16, 2019
Messages
20
Location
Darmstadt - Germany
I just finished this little tool.
quad-link.jpg

There is nothing special to it, so I open sourced it.

Electronics files:
https://1drv.ms/u/s!AmL4KhemsqWQivJILj5mJUTWvd_BaQ?e=3roIFm

Housing files:
https://1drv.ms/u/s!AmL4KhemsqWQivJGEl4mWK1CDSmcDA?e=pEu7CQ

Frontpanel files:
https://1drv.ms/u/s!AmL4KhemsqWQivJFqj6OZk_EL07Fhg?e=WFR7Sf

quad-link-cat-multicore-platinen-set~2.jpg

quad-link-cat-multicore-platinen-set~3.jpg

quad-link-cat-multicore-platinen-set.jpg

Have fun and stay safe  :)
 
Damn!
Sorry, I just realized that I successfully posted this int the wrong channel. :-[
Now I can not delete it...
Maybe an admin can move this thread to the "Drawing Board"?
 
Nice work. If you can get custom cat 5 cables like rat sound, the you really are styling. Dave’s have individual grounds per twisted pair along side the shielded cat5e he uses. All grounds star at the connector.  Good stuff.
 
I know little about cat5 other than it is four twisted pairs. So how do you connect the XLR pins 1 from end to end?

Cheers

Ian
 
ruffrecords said:
I know little about cat5 other than it is four twisted pairs. So how do you connect the XLR pins 1 from end to end?

Cheers

Ian
Throug the cables shield contact. Works fine.

I'm using it for line level signals and since they are balanced you don't really need a ground anyway.
 
juliusbusch said:
Throug the cables shield contact. Works fine.
OK. I did not know shielded versions exist. Is it easy to get cat5 shielded cables and connectors?
I'm using it for line level signals and since they are balanced you don't really need a ground anyway.
Strictly speaking pin 1 is shield not ground but I take your point.

I could really use a pair of these. My 27ft long workshop has a computer and stand alone DAW at one end and my mixer/module test area and speakers at the other end. At the moment I run a pair of 10m XLR cables from end to end but I could really do with a couple more connections. Four tie lines in one cat5 cable with a nice wall mounted box at either end sounds ideal.

Cheers

Ian
 
I did this more than 20 years ago, when I worked for hospital radio.
We had to build a new studio and we needed a lot of cables, but of course: there was no money...
A company had donated us a long cable, that consisted of twisted pairs with a single woven screen.
At first I was a bit worried about crosstalk, but in reality everything worked fine.
No hum or other problems; we used this cable to transport (dynamic) microphone signals from the studio to the mixer, over 15 meters!
 
ruffrecords said:
I could really use a pair of these. My 27ft long workshop has a computer and stand alone DAW at one end and my mixer/module test area and speakers at the other end. At the moment I run a pair of 10m XLR cables from end to end but I could really do with a couple more connections. Four tie lines in one cat5 cable with a nice wall mounted box at either end sounds ideal.
I have eight PCBs left (in case you don't want to manufacture them yourself):
http://diy-audio-tools.de/Quad-Link-Cat-Multicore-Platinen-Set


RuudNL said:
... in reality everything worked fine.
No hum or other problems; we used this cable to transport (dynamic) microphone signals from the studio to the mixer, over 15 meters!
That's so nice to hear  ;D I was worried about the small microphone levels as well, but good to know that this won't be a problem.
 
Redco has this done this:

  https://www.redco.com/Redco-EA-4P-FX-01-Ethernet-to-4-Female-XLR-breakout-cable-1.html
  https://www.redco.com/Redco-EA-4P-MX-01-Ethernet-to-4-Male-XLR-breakout-cable-1.html

Says crosstalk is -106dB. But it depends on load. If you try to drive 600 ohms to 20dBu next to a 10K load, you're gonna get crosstalk. But if it's phantom powered mic signals, it probably works great.

 
juliusbusch said:
I have eight PCBs left (in case you don't want to manufacture them yourself):
http://diy-audio-tools.de/Quad-Link-Cat-Multicore-Platinen-Set

I checked out your web site but it seems the only way to pay is by bank transfer. Paypal would be so much simpler.

Cheers

Ian
 
squarewave said:
Says crosstalk is -106dB. But it depends on load. If you try to drive 600 ohms to 20dBu next to a 10K load, you're gonna get crosstalk. But if it's phantom powered mic signals, it probably works great.
Could you please cite any article or site with info or some calculations about this dependency?
 
juliusbusch said:
Electronics files:
https://1drv.ms/u/s!AmL4KhemsqWQivJILj5mJUTWvd_BaQ?e=3roIFm

Thanks for sharing, but you did traces sub-optimal.  Traces + and - of one XLR should be placed very close one in relation to the other, and there shouldn't be any other signal between them.  You can do that much better. Don't use Autorouter here. NHF please.
 

Attachments

  • 4xlrtocat5.png
    4xlrtocat5.png
    221.5 KB
moamps said:
Could you please cite any article or site with info or some calculations about this dependency?
Nope. Just using logic.

Cat cable does not have shielding between pairs. So there will be electrostatic and electromagnetic crosstalk. Just spit-balling here but I think electrostatic will be dominant if the impedances are high and electromagnetic will be dominant if current is high. I don't think the typical line input impedance of 10K is particularly high wrt capacitive coupling so I think the primary concern would be electromagnetic coupling. In that case, a 600 ohm load would be about as low as one would expect for a line level signal. Lower levels would equate to lower current. So high level signal into 600 ohms running next to a 10K line is probably going to generate about the most electomagentic coupling that you would expect to see in reality.

Now is that crosstalk higher than your noise floor? If you have an otherwise very quiet source running into a converter, then I would be willing to bet a roll of toilet paper that you would see it on an analyzer.

UPDATE: Actually I must be wr-wr-wrong about this. All of the cables Whoops is posting about below use shielding that would only help with capacitive coupled crosstalk. In particular, SFTP is foil around each pair and then foil and braid around the bundle. But I don't think the extra foil would create enough space to reduce electromagnetic coupling. In regular snake cable each pair has a vinyl jacket around it that creates a relatively large amount of space (2x jacket is 2-3mm maybe) between the pairs. It would be interested to measuring the difference in crosstalk between a conventional CAT6 vs SFTP CAT6 vs proper snake cable.
 
john12ax7 said:
Been meaning to try this as well.  Is there a preferred pre-made CAT cable to use for analog audio signals?

Any shielded Cat5 or Cat6 will do.
If it's for fixed installation you don't need flexible cables, so it will be cheaper and easier to find.

If you need the cable to be flexible, then there's a lot of good options from S2CEB, like the Cat6 cable or Audiolan 24 (a bit thick)
I think the S2CEB Cat6 cable has individual shields, although you don't need that.


 

Attachments

  • Screen Shot 2020-05-05 at 1.24.28 AM.png
    Screen Shot 2020-05-05 at 1.24.28 AM.png
    240.3 KB
Thanks juliusbusch for sharing the docs,
but there's really cheap options for this, with these prices I don't see any motivation to DIY one:


qV7Zw1Kl.png

https://www.thomann.de/gb/the_sssnake_cat_snake_3mb.htm



sEdQGt4l.png

https://www.thomann.de/gb/the_sssnake_cat_snake_3fc.htm



nKJg4Owl.png

https://www.thomann.de/gb/stairville_rj45_dmx_split_box_fx4.htm



oGPsgxXl.png

https://www.thomann.de/gb/stairville_rj45_dmx_split_box_mx4.htm


 
Back
Top