common or separated grounds on a sound multicore ?

[Mystic]

I'm currently rewiring a multicore which have been cut, and I'm asking myself, should I patch it with a common ground (all individual channels share the same ground) or do I have to patch it with a separate ground on each channel ?

 

thanks for your help.

[mal421]

I'm currently rewiring a multicore which have been cut, and I'm asking myself, should I patch it with a common ground (all individual channels share the same ground) or do I have to patch it with a separate ground on each channel ?

 

thanks for your help.

 

All the individual grounds will eventually go to a commom ground at the mixing desk . I would do whichever gives the neatest job . However if the individual pairs are foil screened you will have to connect all the drains to earth .

[RonJon]

Individual Screens are absolutly essential, do NOT common the screens under any circamstances. Sorry to come across severe but don't join them up.

 

Reasons:

 

1: Joining the screens will create a large amount of cross talk between the channels, so you will be liable to hear more audible colouration.

 

2: By joining the screens you will have a secondery point of grounding in the middle of the loop, which will effectivly nullify the screening potential of the cables, and distribute any noise on one pair, handily over the entire multicore.

 

3: You also run the risk, depending on how any unbalancers you have work, of firing 48v up the bum of equipment plugged into the system when phantom is live. Can be a problem, tends to rear its head as singers get a gentle smack off a 58 through their lips, amusing for a while, but does tend to get boring..

 

Would recomend joining the whole lot with XLRs, or if your soldering is good, with christmas tress, or if not, then krone block (though can have reliability issues if not connected carefully)in adaptable boxes.

 

Hope that answeres your question.

[Simon Lewis]

RonJon,

 

Could you clarify a couple of points here....

 

1: Joining the screens will create a large amount of cross talk between the channels, so you will be liable to hear more audible colouration.

 

I understood the external noise and crosstalk rejection to be primarily a function of the conductor pair twist and lay length. The screen is effective at RF but the cable twist gives the greatest immunity to LF interference.

 

Granted, there is a difference between foil screen & drain wire construction and braided construction (the braided types give superior RF protection and help reduce shield current induced noise). I also accept that a break in the cable twist will potentially allow noise to enter the cable system. However, I'm not fully convinced that the crosstalk arguement is correct? Given that the shields are not part of the signal chain (in a telescoped ground system, they are disconnected at one end) I wouldn't expect crosstalk per se to be such an issue. There is an arguement regarding the shield current of one circuit being imposed on other circuits, but in a well behaved system with correct Pin 1 connection that shouldn't be such a problem (!). Perhaps this is the issue you had in mind?

 

2: By joining the screens you will have a secondery point of grounding in the middle of the loop, which will effectivly nullify the screening potential of the cables, and distribute any noise on one pair, handily over the entire multicore.

I think this is is my point above.... a problem in one circuit is "shared" across the shields of others. However, is this actually a secondary grounding point?

 

3: You also run the risk, depending on how any unbalancers you have work, of firing 48v up the bum of equipment plugged into the system when phantom is live. Can be a problem, tends to rear its head as singers get a gentle smack off a 58 through their lips, amusing for a while, but does tend to get boring..

 

yep.. although the phantom power supply is current limited and is driving into a short circuit...

 

 

I'm not necessarily disagreeing with you, it's just the terminology didn't look to be correct!

 

Another problem is using th audio multicore to carry Tecpro comms circuits etc. which may use the screen as a signalling conductor. In such cases, separate screens are quite important!

 

So, OP, keep your grounds separate, and maintain maximum twist up to the point where the cables are joined!

 

 

Simon

[Jordan]

Hi all, I look after my bands PA. A task I enjoy, and although I really have very little knowledge of sound engineering I am learning all the time and enjoying the process.

 

I recently bought a length of Belden multicore from ebay with edac connectors at each end. The first thing I did was check all the cores for continuity and all checked out OK. Next I checked to make sure none of the cores had continuity with any of the other cores and found that two of the cores were indeed contacting each other and also contacting the centre screw of the edac. I opened the housings and found that the two cores that were making contact were actually two uninsulated ground cores and that the centrescrew had actually rubbed through the foil screen on two chanels thus contacting the cores.

 

I think this should be easilly remedied with tape but my question is, does it actually matter if two ground cores from two different channels make contact. Thanks

 

 

This post merged in from another topic. Before posting a question, would members please remember to search the archives to see if a similar question already exists? Thank you.

[Simon Lewis]

Jordan,

 

In theory this should still work OK, since the multicore ground drain wires will all terminate to a common point in the mixing desk.

 

However, various manufacturers have employed different grounding methods for their equipment, mainly to do with whether the shield terminates at the equipment enclosure (i.e. is joined to the mains earth) or to the circuit board (where it may join the audio 0V).

 

This is a problem that has caused all sorts of hum issues, and is still a hot potato...

The upshot is, depending upon how your equipment is wired, having multicore shields touching each other could possibly add noise to your system, but in a well designed system, it shouldn't. What happens when you wire it all up?

 

If you want to read more on this topic, click here and here...

 

HTH,

 

Simon