This comment pertains connections where no shielded enclosure exists.
4.3 outlines what to do when the mechanical enclosure is not an effective shield (e.g. made of plastic). In its present state it covers only the case where PCB mounting connectors are used. In the case of wired connectors, the point where the shield is separated from the signal wires becomes arbitrary.
It is better to continue the shield around the twisted pair (ie. solder a shielded TP cable to the connector) up to the PCB and separating the shield from the signal wires there (see (1) on attached drawing).
Better still would be to run an STP cable from the chassis connector to the Star Point, terminating the shield to the SP on the spot and continuing the two signal wires from there (second drawing).
"When the connector is wired to the signal circuitry, the connection shall be made using a shielded cable and the shield terminated directly to the star point. The signal wires can continue to their destination from there."
The suggestion that the cable should be taken as close to the signal circuitry as possible before the connection of the cable shield to the star point is made is in complete accordance with 4.3. The arrangement shown in Figure 3 includes a notional outline for an enclosure that has no shielding properties and is included for comparison with other figures that do have shielding enclosures. The location of the connectors with regard to this non-shielding enclosure is not electrically significant.
Because the only shielding available in this configuration is in the cable shield itself, it clearly makes good engineering sense to continue this shield as close to the signal circuitry as possible while maintaining the essential topology shown in Figure 3. It is proposed to add an explanatory note as follows:
"NOTE: Good engineering practice encourages implementers to locate the 'Star Point', the connectors, and the signal circuitry as close as possible to each other in order to gain the greatest benefit from the cable shielding."
I believe that section 4.5 "Shield Interruptions" should be removed. This paragraph undermines the whole ground mesh methodology of IEC 6100-5-2. It implies the existence of good reasons to interrupt the shield connection without any explicit justification.
The reference to A.6 is misleading. It starts with the premise that one end of the shield will be disconnected to reduce the effect of cable capacitance mismatch and then recommends which end is best. Fair enough but not relevant.
We can assumes that modern equipment's low driver output impedance will dilute the effect of cable capacitance unbalance. Furthermore we must take into account the danger of flashover from lightning activity* and all the present day EMC issues (mobile phones and digital audio circuitry).
No published AES document should even implicitly suggest that there is no better alternative technical solution than to break a shield.
Enrico Cecconi M.Sc.
I believe that there is nothing in clause 4.5 that is in conflict with IEC 61000-5-2. Any grounding scheme can be used with any of the shielding schemes defined in AES48.
Draft AES48-xxxx is scoped to consider only the connection of the designated shield contact within equipment. It does not extend to the wiring of the building which should, of course, follow the regulations and best practice applicable for that building.
The scope of the IEC Technical Report IEC 61000-5-2 "Electromagnetic compatibility (EMC) - Part 5: Installation and mitigation guidelines - Section 2: Earthing and cabling" (Ed.1, 1997-11) makes clear that, "It applies primarily to new installations, but where economically feasible, it may be applied to extensions or modifications to existing facilities".
IEC 61000-5-2, 6.4.2, points out, "Shields of cables are bonded to the earthing network at one or two extremities depending on the signals being transmitted and on possible electromagnetic interference sources."
The purpose of AES48 4.5 is to deprecate the current practice by some manufacturers of making no connection to pin 1 at all inside equipment. The AES Standards working group views this practice as seriously misguided.
As IEC 61000-5-2, 7.4, also recommends, "In shielded bifilar or multi-lead cables, the shield should be regarded as a PEC (Parallel Earthing Conductor)."
Nothing in AES48 contradicts this recommendation. However, in the case of electrical installations in older buildings that may not follow the guidelines in IEC 61000-5-2, for example, this approach may not always be practical or prudent - hence the wording in 4.5. In such cases it may be appropriate to apply a dedicated PEC, however this is outside the necessarily limited scope of this document.