Recirculating currents?

[powrgen]

Can anyone explain the concept of recirculating earth currents please? It is our practice to earth at one end -or at a single point only, in a distribution system to 'avoid recirculating currents' We have blindly followed this practice without giving it too much thought but now I'd like to know more..
Thanks
PG

 

[gordonl]

Circulating currents are induced in the shields of shielded power cables due to the flow of current in the conductor. Similar effect occurs in shielded control cables due to noise If the cable shield is grounded at both ends the induced current will flow in a loop through the shield and return through the ground path. The longer the run of cable the higher this circulating current is, this circulating current heats the cable and can reduce the ampacity of the run. To avoid this problem one end of the cable run is left open so the current doesn't have a path to flow through, but care must be taken because the ungrounded end of the cable can have a large induced voltage which can be dangerous to people and equipement.

To avoid the overvoltage situation we've installed surge arrestors on the ungrounded end of long runs.

Also see the following excerpt from Southwire:

http://www.southwire.com/tech/library/pc/pwrcab04.htm

 

[peterb]

Just to add a word to Gordonl's post -
When a system ground fault occurs, the current flowing in the earth results in different potentials at different locations. If you connect separate ends of a cable shield to ground, this creates a parallel path for ground current to flow, giving rise to the effects noted above.
For single end grounding, the shield transfers the potential of the grounded end to the other end, giving rise to the need for surge arrestors and various other means of personnel/equipment protection. In some cases, the shield is grounded at the middle of the cable run to mitigate this condition.

On re-reading the original post, it seems that Powergen may have meant system grounding, rather than cable grounding. If this was the case, then the answer is that single point grounding makes ground fault detection and protection somewhat easier, as all ground fault current returns through the grounded transformer neutral. Multipoint grounding is also widely used in distribution systems, as in the multi-grounded common neutral configuration. For this case, some of the ground fault current returns to the source transformer neutral via the system neutral conductor, as does unbalanced load current.

 

[powrgen]

Thanks chaps

 

[jbartos]

Suggestion: Both ends grounding is sometimes done if the circuit demands it. Specifically, if the ungrounded end raised potential causes interference with sensitive circuit, e.g. in instrumentation. Then, both ends are grounded, bringing the highest raised voltage in the middle of the run and both ends are at the zero potential. It is to be noted that this is usually done for applications where any circulated currents are negligible or small (could be dc).

 

[buzzp]

Basically, if you have multiple grounds then these grounds will be at slightly different(or significant) potentials. If you could, you could measure with a volt meter between the two different grounds and you would measure a voltage. This can have dramatic effects especially for circuits that are doing any measuring (and using the 'ground' as the reference point). In general, multiple 'grounds' should not be used unless hazards will exist that someone has already talked about. Buzzp

 

[electricpete]

Double-ended grounding is sometimes done in ac power circuits, but never in control/signla circuits.

From a magnetic shielding standpoint, a double-end grounded cable will have a ground current that tends to flow in opposition to the net conductor current, providing a self-magnetic shielding effect against interfering with nearby sensitive circuits. But this circulating current also causes heating which must be accounted for in establishing the cable current-carrying capability.

Also the double-end grounded cable avoids the concern of a large induced voltage which may exist at the ungrounded end of long single-end grounded cables.

As far as how the doulbe-end grounded cable behaves during a fault which might tend to establish differences in ground potential at both ends... I don't know.