View Single Post
Old 3rd July 2020
Lives for gear
David Rick's Avatar
The answer to many of these questions depends on what kind of interference challenge you're trying to mitigate or avoid. A foil shield provides 100% coverage, as opposed to 75-95% for a braided shield. This provides a shielding advantage for very high frequency electric fields, but it comes at the cost of higher shield resistance, which causes a disadvantage for lower frequency interference. Cable manufacturers typically add a drain wire for ease of termination. Although this lowers the shield resistance at low frequencies, it's also known to reduce common-mode rejection of inductive pickup because the current induced in the shield flows asymmetrically (mostly in the drain wire) and this reduces the effective balancing of the twisted pairs you're trying to protect. The other caution about foil shielded cable is that it's only appropriate for stationary installation, because it won't withstand frequent flexing. Spiral "served" shields can cause similar problems current distribution problems, which some manufacturers try to mitigate by making the shield spiral the opposite direction of the pair twists. Such a shield tends to have higher inductance than a braided shield, as well. The relative ease of termination compared to braided shields is a big attraction however. But a disadvantage is that the shield can develop gaps if the cable is abused, say, by rolling a piano over it. For really critical applications, there exist cables available that have both braid and foil shielding. These are absolute murder to terminate, so most people avoid them unless they're truly needed.

My recommendation is to clearly understand the particular RF challenges your installation faces (not forgetting handheld personal devices) and make your cable choice appropriately. How long are the cable runs? Is there low impedance technical grounding to prevent ground-difference currents circulating in the audio shields? The choice of whether to connect one or both ends of the shields rests mostly on this. But where there is potential interference such that the cable length becomes a significant fraction of the interfering wavelength, then capacitive coupling of the "lifted" end is recommended. The exact capacitance value isn't too critical, but it must have low impedance at the challenge frequency, and high impedance at power line frequency. It should also be large compared to the total (i.e. full length) shield to signal wire capacitance.

David L. Rick