Cross Bonding Verification Testing 1

[WhiteyWhitey]

Hiya,

I am commission a 275kV underground cable system.

System:
Conductor resistance is about 15m? per phase.
Each phase is a single core with Aluminium corrugated sheath cross bonded at the joint bays.

Background:
I am trying to test the cross bonding system. The goal being to prove using an onsite test that the cross bonding system is correctly connected and sufficiently cancels sheath currents.

The process I have been given is something like this:

1. Connect 3 phases of the remote end of the feeder together and to ground.

2. Connect a 3 phase 50A (thats Minimum, Preferably would like 100A) source to the feeder.

3. Inject 50A into the main conductor

4. At the joint bay record sheath induced curent and voltage on each sheath.

5. Reverse the cross bonding

6. Repeat measurement.

The theory is you should get current induced in the sheath of about 10A when correctly connected and this will be higher if the cross bonding is reversed (Say 30A)


Problem:
The machine I have is designed more for relay testing than current injection. The test set can produce 200A for 1 min when given a return current path (it has A phase output and A phase Common). When I maxed out the machine on the conductor I could only get it to produce 15A.

The measurements at the joint bay are too inaccurate and variable to be usable.

Question:

Does anyone have experience with this test?
What equipment should be used?
Is there a better way to test the cross bonding system?

Thanks in advance for your always Awesome tech advice.

Whitey.

 

[ausphil]

Whitey,

We perform this test on all our 132kV cables as part of the commissioning program prior to energisation. It has picked up a number of occasions where the cross bonding links were put in incorrectly, and were rectified. The currents and voltages at the cross bonding boxes when the system is connected correctly depends on the current that you inject into the cores, and the design lengths of the cross bonding sections. The currents and voltages you get when the system is not connected correctly would generally be significantly larger than the correct ones (that is it may not be 10A and 30A).

We make one change to the cross bonding arrangement per cross bonding group (ie 3 sheaths sections), not completely reverse the connections. If you have 4 cross bonding groups (12 sheath sections) you would end up doing 4 individual changes and rectifications. Also, after you return the connections to normal, it is best to redo your measurements to ensure that it has been restored correctly.

Ultimately, this is the only real way to test your cross bonding system. And the beauty of it is that you can measure your positive phase sequence impedance in the same test setup. Plus with some setup changes, you can measure the zero sequence impedance as well.

My suggestion is not to connect the remote end connections to ground, just connect the 3 phases together. This eliminates any currents flowing in the ground or places that you don't want them. It is just a balanced 3 phase injection.

The equipment we use is three 50A variacs (variable output autotransformers) to control the input voltage to a 415-20V supply transformer. This steps the voltage down, and gets the current rating up to what we require. We aim for 100A, but have accepted about 20A in some circumstances. This will depend on whether you can get a good reading from the test, which is mainly due to any standing (background) voltages or currents in the screens prior to applying your test voltage.

The step that you didn't mention is to scale the current values up to the rated current for the cable system. This will give you the currents that will be flowing in the screens at full load. Depending on the values, the cable may need to be derated due to excessive screen currents. You also have to scale the voltage values up too.

In summary, a test well worth doing, especially at the higher voltages, because it is an asset that is going to be in service for 50 years (plus), and to start it behind the 8-ball by having an incorrect link arrangement, and hence higher screen currents day in day out, as well as higher than designed fault voltages on your sheath, is not a good way to kick it off.

Also, don't forget to measure currents in earth continuity conductors if it is an unbalanced arrangement, and voltages on the sheaths at your sealing ends (or whatever cable termination you have) if it is not an earth point.

Ausphil

 

[WhiteyWhitey]

Ausphil,

Thanks for the great response.

We currently have the T&R 200A 3Ph/ED test set available.
However, I have Reservations as to if this set is able to produce enough current without a current return to the machine.

Do you have any contacts in Australia that are able to perform the test or hire out variacs and step down Tx's?

Also,
What modifications are needed for measurment of positive and zero sequence impedance?


Thanks Again.

Andrew

 

[ausphil]

Whitey,

If the test set is 3 phase, then by connecting the phases together at the remote end, you will have your current return when the system is balanced.

To measure positive sequence impedance, you only need to measure the voltages, currents and phase angles in this configuration, and calculate the impedances, or use a meter that does the calculation for you, or the test set probably calculates it.

To measure zero sequence impedance, you need to connect the phases together at the testing end, and supply a single phase voltage between this connection and earth. At the far end, you need to connect the phases together and earth them as well. The zero phase sequence impedance looks at the return path through cable sheaths and the earth to the testing end.

There would be many companies that can perform the tests, it really depends on where you are. Most distribution authorities would be able to perform the tests. if you are going to be doing a lot of tests, then you would be best to but the equipment, but if only for a couple of tests, then get a test company to do it.

Most power system books have the equivalent circuits for positive and zero sequence impendances. Basically a zero sequence network is used for analysing a phase to earth fault, and the positive sequence is used for analysing normal balanced operation or symmetrical 3 phase faults, so you should be able to see how the test setup is measuring a particular impedance.


ausphil.