Armor current 1

[jimmy2times]

This relates to another recent thread i posted 238-414369

We measured circ current in installation with single corre cables laid in non-trefoil arrangement for 20% of cable run. Breaks out of trefoil then in to groups of three parallel phase conductors per phase with approx 6" between phase groupings at either end of cable run. Circ current measured 130A at part load. Disconnected transformer end ground braiding by disconnecting bolt connection to ground bar and insulating lug end (ground braiding is insulated itself). Cable armour and cable screen individually insulated. Converter end remained grounded

Was end expecting to see little current. Still measured 15A under similar part load? An order lower but expected 1A max. Any ideas?

 

[7anoter4]

In my opinion, if the armor is grounded at a single end and no connection between single-core cable armors exists no circulating current could flow through armor. However a built-up induced voltage it could be measured between the other end and ground.
In order to calculate this voltage see:
IEEE 575 IEEE Guide for the Application of Sheath-Bonding Methods for Single-Conductor Cables and the Calculation of Induced Voltages and Currents in Cable Sheaths. Appendix D Calculation of Induced Voltages.
For two parallel cable per phase see also:
EPRI-EL-5036-V4 page-4-15 Table 4-6 or:
Underground System Reference Book -1957 ch.10 Table 26.

 

[jimmy2times]

Have looked at that standard (i have an out of date copy) and it calculates maybe 4V at unearthed end (only 22m cable length, 980A).

Only thing i could think of was that the "isolated" braids were still sat on the metalwork on the transformer terminal box during the test (not cut back to the screen termination point). So the cable braids (which were insulated) in close proximity to earth metalwork. Whether there could still be inductive current as result of this? No possibility to repeat test at the moment so thinking about the result and trying to understand why current would not be near zero.

 

[waross]

Metering error?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jimmy2times]

Placed a spare current probe in compartment hanging in free air within cabinet to check if spurious reading and that read near zero during same test.

Four different analysers looking at different braids all reading similar

 

[jimmy2times]

Still struggling to understand currents we measure with single end bond of cable screen as dicussed above. OEM of drive took indepedent measurements and had same findings, there was a clear impression of pulse waveform of the converter current in the screen current. Could this be cable eddy current that we are measuring at screen bonded end? the cable run is far from perfect and along the route where it is trefoil the groups of the adacent trefoils compromise the 1d cable spacing rule.

 

[submonkey]

Hi,

Is there any possibility that the screens were not connected
to earth, but were connected to each other? Anything which
forms a conductive loop linked by flux from the phases will
have current in it. How and where did you measure the circulating
current? Could it be cable charging current? Does it vary
proportionally with load?

Thanks,
Alan

 

[jimmy2times]

Cable is single core with non magnetic screen and armour. 3 cables (240mm.sq) per phase. 12 pulse transformer so 18 cables total betwen tx and converter, 9 for delta and 9 for star windings. Transformer end screen and armour isolated now (from each other and earth), previosuly both were separately bonded to local earth bar in TX. So all 36 earths braids for the cable screen and armour met at local earth bar. The arrangement at the converter end is similar, This remains in situation where the cable armour and screen are separately bonded to local earth bar via flat HF braids 18 off. The braids themselves are insulated, look photos at linked thread in OP to see get a better picture.

We are now measuring at converter end, separately looking at cable armour and screen. We have monitored all 6 braids (sreen and armour times 3) of R phase of delta winding and all 6 of star winding.

Have now made further testing at full power (result in OP was from 20% load) and have made a test where we looked at existing double end bonded and new single point bond arrangement.

For double end bonded at max load we were seeing approx 75-135A on all 6 braids of one phase. Single point bonded seeing 55-70A on one of the 6 braids, next highest around 30-40A then low teens on others so not all 6 braids of same phase consistently high as per double end bonded arrangement.

Current is proportional to load.

Planning now to isolate the local earth bar at converter end from earth frame and have just one earth connection to the earth frame such that we can monitor what the combined effect of all 36 braids to earth is.