Allowable cable ground currents

[shake1]

If in the normal day to day scheme of things you decide to measure the ground current in a feeder cable. Lets say a 3 phase 5 KV CLX non shielded cable. What amp reading would begin to raise concerns, 1/2 amp, 1 amp, 2 amps or say just any ampere reading only.
The readings are to be taken with a clamp on ground CT meter.

Comments

 

[jbartos]

Suggestion: It appears that the sought current value somewhat varies with the feeder load size. E.g. 5kV 100kVA load may have a different ground fault return current than 20MVA load.

 

[alehman]

I'm not clear on how you plan to approach this. Where are you going to place the CT for measuring ground current?

Are you doing this because you suspect an insulation problem with the cable? I would think the charging capacitance of the cable will mask any small level leakage to ground. Any level of true leakage current over a few milliamps per 1000 ft would warrent investigation.

I would suggest a DC high-potential as a much better means of evaluating cable insulation.

 

[shake1]

Testing some old cables at 13.8KV, & 4160. On the 13.8 finding Zero to 1.3 amps. On the 4160 finding Zero to 2 amps except for one at 13. In the process of installing a replacement for the 13 amp find for when it blows or when I can take it down in several months.

The cables are in the range of 35 - 40 years old. Would not even consider trying to hi-pot them unless I had a replacement installed and ready to go into service.

Partial discharge testing on the cable with 13 amp reading showed nothing, other words it is in continous discharge.

 

[busbar]

shake1 — Are the clamp-on ground CT currents reflected as residual or zero-sequence {ground-sensor} currents in associated protective relaying? How do they compare with in-service relay pickup settings?

 

[alehman]

I'm still not clear where you measured these currents.
If you can respond to busbar's questions that would be helpful. To word it another way, do they represent current in the neutral to ground connection at the source or was your CT surrounding all three cables? If it is ground current, is anything else connected to the source?

Were the cables connected to anything other than the source when these readings were made? How long are the cables? Are the 13kV cables also unshielded?

I agree that you should not hi-pot 35-40 year old cables unless there has been regular maintenance testing.

 

[shake1]

The readings were made with a hand held clamp-on ammeter used primarily for locating ground faults on our system, meter is calibrated yearly by outside firm. The meter was clamped over a 3 phase CLX cable and then read. The cables range in length from 350 ft to 1200 ft. The one in question (13 amp reading) is one of a paralell 500 MCM feeder approximately 550 ft long, normal load on this service is about 500 amps. The 5KV lines are non-shielded, the 13.8 KV lines are shielded.

The cables were in normal service when metered.

The cable with the 13 amp reading has a PJC11 set at 200 amps to clear. Alas the ground CT's do not have an amp meter for display. This is on a main sub, the utility sub feeding this main sub is set at 400 amps to clear. Low impedance ground on this service.

Why the high (seemingly) settings? I am still investigating old works.

 

[busbar]

Offhand, 13 amps in 500 may not be a significant cause for concern, except maybe to check periodically for changes.

Is it possible to use an XLA13 paddle and meter at the PJC11 to eliminate concerns there?

 

[DanDel]

shake1, very likely you are measuring circulating ground current in the cable shield caused by induction.
Is the system solidly grounded or grounded through a resistor or reactor?

 

[busbar]

Agggh! Sorry. XLA12!

 

[alehman]

As dandel suggested - in the non-shielded cable you are likely seeing stray current in the metal jacket due to differing ground potentials at the points where the jacket is grounded and/or current induced by the current in the cables. Is there load on the cables? A few amps wouldn't concern me much. It may also be due to some type of non-symmetry preventing perfect cancellation of the magnetic field where you are measuring. I don't think it's an insulation problem if your PD tests were negative.

If there are parallel cables, it wouldn't surprise me to see more current due to the possbiltiy of circulating current. 13 amps might be worth investigating. A difference between two parallel cables might be due different effectiveness of the jacket grounding.

For shielded cables you would expect to see current induced in the shield if it is grounded at more than one location. The only way to negate this is to run the shield drain wire back through the CT.

 

[jbartos]

Suggestion: The cables have available capacitance value per length that is used for calculation charging current. Charging currents are used in grounding calculations to determine a system grounding type, e.g. high-resistance system grounding, medium-resistance systems grounding, etc.
What is being referred to as stray current is the cable and potentially loads capacitive charging currents.

 

[alehman]

Charging current could certainly be a contributor in the case of the non-shielded cables. In shielded cables, charging current should flow to the shield, not the jacket.