Not certain what you mean by "this".billshurv said:
If both ends have a PE connection then some audio return current will return via that connection, unless the audio circuit is isolated at least at one end. You can reduce the amount of audio choosing that route by reducing the resistance of the intended audio return connection; using one wire of a twisted pair as the return does the opposite as it will be much higher in resistance than a coax shield.
When reading anything on cables, shielding etc. always ask the question: is this about audio or RF? Lessons learned in one application do not transfer very well to the other.
For clarity, its important to separate LF ground (or noise) loops from HF (i.e. RF) noise loops.
A cable shield does nothing at LF and the only mitigation available in that case is keeping the loop area between the return and the signal carrying wire as small as possible, which is what you get with a coax cable anyway. In a balanced system at RF, both ends of the shield are connected to their corresponding chassis to the whole thing forms a single faraday cage.
You will have LF magnetic flux coupling into the loop bounded by the shield, the PE connections in the amplifier and back to the power sockets. But, if you have a balanced system, the common mode noise is rejected and you have no problem.
On single ended systems, I normally do not connect the shield to the chassis because you then get a LF noise loop formed between the point the shield connects to the chassis (i.e. at the input connector) and the power earth connection to the chassis and the PSU 0V inside the amplifier. However, if you wire a small cap of a few nF from the shield to the chassis at the point of entry, it is a short at RF and you have a faraday shield providing some protection from RF ingress. The best solution for LF noise in single ended systems are ground lifters (with lots of safety caveats), HBR's to reduce cross channel loops and a relentless focus on reducing loop areas both inside the amplifier and between equipment.
A cable shield does nothing at LF and the only mitigation available in that case is keeping the loop area between the return and the signal carrying wire as small as possible, which is what you get with a coax cable anyway. In a balanced system at RF, both ends of the shield are connected to their corresponding chassis to the whole thing forms a single faraday cage.
You will have LF magnetic flux coupling into the loop bounded by the shield, the PE connections in the amplifier and back to the power sockets. But, if you have a balanced system, the common mode noise is rejected and you have no problem.
On single ended systems, I normally do not connect the shield to the chassis because you then get a LF noise loop formed between the point the shield connects to the chassis (i.e. at the input connector) and the power earth connection to the chassis and the PSU 0V inside the amplifier. However, if you wire a small cap of a few nF from the shield to the chassis at the point of entry, it is a short at RF and you have a faraday shield providing some protection from RF ingress. The best solution for LF noise in single ended systems are ground lifters (with lots of safety caveats), HBR's to reduce cross channel loops and a relentless focus on reducing loop areas both inside the amplifier and between equipment.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.