Beginner question on mounting RCA jack in chassis

Status
Not open for further replies.
Jump to Latest
#2
They should be isolated to prevent ground loops and power supply 'noise' entering the audio chain.

Isolated sockets are readily available and come with a thin plastic spacer and washer per socket.
 

Attachments

  • Capture1.PNG
    Capture1.PNG
    749.6 KB · Views: 599
#3
This needs to be considered when planning the grounding scheme. Commercial equipment usually chooses a star ground point at the plate with firmly connected non-insulated jacks these days - which is (a) technically the better option and (b) just about the only thing you can do when you need a gazillion jacks without intending to go crazy.
 
#5
Some audio circuit designers say that they should be isolated.
But the experts like:
Neil Muncy (RIP)
Henry Ott
Jim Brown
Bill Whitlock
Think that the shield of all chassis connectors should be connected to the chassis.

*******************
*******************
And no a capacitor to the chassis doesn't make it correct.
 
#7
Well... Here's some links...
http://www.hottconsultants.com/pdf_files/aes-2007.pdf
http://www.hottconsultants.com/pdf_files/Audio Interconnections.pdf
Grounding and Shielding Audio Devices
http://audiosystemsgroup.com/IEEE-Tutorial05.pdf
https://centralindianaaes.files.wordpress.com/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf
http://jensen-transformers.com/wp-content/uploads/2014/09/Audio-Transformers-Chapter.pdf
https://shure.custhelp.com/ci/fattach/get/97648/0/filename/Ground+Loop+_Bill+Whitlock.pdf
https://www.youtube.com/watch?v=g9M45U87ybg
http://espsurgex.com/downloads/white-papers/PowerGround.pdf


My thoughts are it would be worth collecting a range of relevant links as a easy to reference list as this question comes up often and the solution is not always easy..., I have more but they are on my work system.
 
#9
Speedskater said:
Some audio circuit designers say that they should be isolated.
But the experts like:
Neil Muncy (RIP)
Henry Ott
Jim Brown
Bill Whitlock
Think that the shield of all chassis connectors should be connected to the chassis.

*******************
*******************
And no a capacitor to the chassis doesn't make it correct.
Click to expand...
The "But" sounds like the list of experts disagree with the opening statement.
I don't think that Muncy, Ott, Brown nor Whitlock disagree. I think they all said that the signal lines should be isolated from the chassis and the SCREEN/SHIELD around the signal lines should connect the chassis together.
 
Last edited:
#14
Speedskater said:
Some audio circuit designers say that they should be isolated.
But the experts like:
Neil Muncy (RIP)
Henry Ott
Jim Brown
Bill Whitlock
Think that the shield of all chassis connectors should be connected to the chassis.

*******************
*******************
And no a capacitor to the chassis doesn't make it correct.
Click to expand...

AndrewT said:
The "But" sounds like the list of experts disagree with the opening statement.
I don't think that Muncy, Ott, Brown nor Whitlock disagree. I think they all said that the signal lines should be isolated from the chassis and the SCREEN/SHIELD around the signal lines should connect the chassis together.
Click to expand...

Speedskater said:
That's hard to do with an RCA unbalanced interconnect.
Click to expand...
an unbalanced 2core, or coaxial, uses two signal wires. Both should not be connected to chassis. A coaxial does not have an exclusive screen/shield. The outer is a signal conductor.
It's the screen/shield around the signal wires that should be connected to the chassis. All your listed Authors agree on this wiring method.
 
#15
forr said:
Data may be more convincing than words :

http://www.diyaudio.com/forums/anal...nal-grounds-chassis-vs-pcb-5.html#post4816268

The conclusion ot the post is : results are clearly better [...] when all ground are connected to the box in order to benefit of the steel box shielding and of the low impedance ground path. Quite curiously, it was not commented and the thread discontinued.
Click to expand...
the result for the two test measurements is clearly different and clearly shows an advantage to the everything connected to the enclosure.
This particular result is for a single channel with an input line shorted to the enclosure.
Unfortunately we don't have the test for two signal wires coming from a remote source, nor for two channels coming from a remote source. These last two are how the RIAA pre would be used in real life.
I suspect his "everything connected to enclosure" will show a difference cf the norm of one connection to enclosure.
 
#16
forr said:
Data may be more convincing than words :

http://www.diyaudio.com/forums/anal...nal-grounds-chassis-vs-pcb-5.html#post4816268

The conclusion ot the post is : results are clearly better [...] when all ground are connected to the box in order to benefit of the steel box shielding and of the low impedance ground path. Quite curiously, it was not commented and the thread discontinued.
Click to expand...
He was comparing two unusual situations:
1. multiple connections from signal grounds to the chassis;
2. no connections from signal grounds to the chassis.
Both might be expected to give poorer results than a good grounding scheme, so it is perhaps pot luck which one turned out to be better. He should have also tried
3. one connection from signal ground to chassis.
 
#17
Speedskater said:
Maybe but I don't recall Jim Brown, who is always worried about RFI ever suggesting it.
Click to expand...

Others have though, there are many points of view and information regarding this, that is why I posted an eclectic list, I have many many more related links on my work system, as I am off sick I cannot get to them at the moment.
Basically the cap creates a low impedance path for high frequency RF noise picked up on the screen, if the screen is not connected at both ends then it is not a screen and will not screen any RF, in fact even worse than that it can become an antenna and pick up noise...
 
#18
AndrewT said:
an unbalanced 2core, or coaxial, uses two signal wires. Both should not be connected to chassis. A coaxial does not have an exclusive screen/shield. The outer is a signal conductor.
It's the screen/shield around the signal wires that should be connected to the chassis. All your listed Authors agree on this wiring method.
Click to expand...

This pseudo balanced routing may or may not work depending on the routing on the relevant PCBs etc. and in some cases it can cause problems. there is no magic cure unless you use fully balanced kit... Then you can worry about pin 1 problems🙂
 
#19
AndrewT said:
the result for the two test measurements is clearly different and clearly shows an advantage to the everything connected to the enclosure.This particular result is for a single channel with an input line shorted to the enclosure.
Unfortunately we don't have the test for two signal wires coming from a remote source, nor for two channels coming from a remote source. These last two are how the RIAA pre would be used in real life.
I suspect his "everything connected to enclosure" will show a difference cf the norm of one connection to enclosure.
Click to expand...
A good two channels RCA connection needs to have the two screened cables maintained as parallel as possible, avoiding the creation of a ground loop. As soon as the cable presents a kind of snake tongue as below, the signals are prone to degradations.

9251826_800.jpg
 
#20
forr said:
A good two channels RCA connection needs to have the two screened cables maintained as parallel as possible, avoiding the creation of a ground loop. As soon as the cable presents a kind of snake tongue as below, the signals are prone to degradations.

9251826_800.jpg
Click to expand...
agreed that lessening the loop area is good.
Even better might be a figure of 8 coax where there is electrical contact.
But would electrical contact degrade the effectiveness of the screening?
 
Status
Not open for further replies.
Top Bottom