3 Phase HV Cable Sheath Cross Bonding 2

[stanyadji]

As we know under ground HV Cable sheath bonding methode are:
single point bonding and cross bonding methode.
With single bonding we have to derated the current carrying capacity of the cable. That's why a lot of engineers prefer cross bonding methode.
I would like to know if the conductor have been transposed at a certain length of cable do we still have to cross bonding the sheath to be able the get the same result?
Best regards,
Stanyadji

 

[jghrist]

Single point bonding does not require derating the cable, but results in a high sheath voltage on the unbonded end. Bonding both ends results in induced sheath currents and the need to derate the cable.

Transposing the cables does not reduce the need for cross bonding. Induced currents or voltage is related to the proximity of the cable sheath to a single phase conductor. This doesn't change with transposing the cable.

 

[HSKabel]

Answer is totally correct. Transposition of phases has nothing to do with elimination of circulating sheath currents that would derate the cable but is sometimes used to balance the phase characteristic impedances for long cable circuits.

 

[submonkey]

Hi,

The two responses above surprise me. I think you would not need (or indeed want) to crossbond cable sheaths if the phase conductors are fully transposed.

My reasoning is described below without a great deal of rigor. If people disagree, perhaps I'll try harder!

Suppose we look at the cable end-on:
[tt]
(A) x (B) (C)
[/tt]
Where (A), (B), (C) represent the phase conductors, and x is a point between two phases, in the same horizontal plane. Assume that all current flows into the page, which means that each conductor produces circular lines of flux which flow in a clockwise direction. At point x, the lines of flux from all phase conductors are vertical and can be summed algebraically. The magnetic field at x due to the current in A-phase is k1*Ia, the magnetic field due to B-phase is k2*Ib, and the magnetic field due to C-phase is k3*Ic where k1, k2, k3 are constants.

The total magnetic field at x is:

Bx = k1*Ia + k2*Ib + k3*Ic

For a balanced three phase system Ia + Ib + Ic = 0, however, k1, k2 and k3 are different, because x is not the same distance from all conductors. An non-zero magnetic field
therefore exists at x.

The cable sheaths form loops which are shown below:
[tt]
+-----------A phase sheath-------------+
| A-C loop x |
+-----------B phase sheath-------------+
| B-C loop |
+-----------C phase sheath-------------+

^^ ^^
Local end Remote end
[/tt]

CASE 1 - conductors not transposed:

The field at point x will be the same when x is shifted along the cable lengthwise. A total flux cuts the A-C loop which induces a voltage and causes current to circulate in the sheaths.

CASE 2 - conductors fully transposed:

We now need to pick three x points, since three cable arrangements exist.

For the first third of the cable run:
[tt]
(A) x (B) (C)
[/tt]
Bx_first_third = k1*Ia + k2*Ib + k3*Ic

For the second third of the cable run:
[tt]
(B) x (C) (A)
[/tt]
Bx_second_third = k1*Ib + k2*Ic + k3*Ia

For the final third of the cable run:
[tt]
(C) x (A) (B)
[/tt]
Bx_final_third = k1*Ic + k2*Ia + k3*Ib

If we sum the magnetic field at the three x points:

Bx_first_third + Bx_second_third + Bx_final_third =
k1*Ia + k2*Ib + k3*Ic +
k1*Ib + k2*Ic + k3*Ia +
k1*Ic + k2*Ia + k3*Ib +

which is equal to zero, because Ia + Ib + Ic = 0.

We can see that for every x point along the cable run, there will be two other x points which provide a cancelling magnetic field. The total flux in the sheath loop is zero, and no circulating current flows.

Am I correct?

Thanks,
submonkey

 

[HSKabel]

Dear Submonkey,
You went through a lot of work to show that the sum of all induced currents in a fictitious conductor in position x is zero, under the condition that phases a,b and c are balanced and the sections of the system are of equal length. So far so good. The sheath or screen of any cable, however, follows its phase conductor and is not in position x along the way. That IS the reason why cross-bonding is needed, so that the sheaths are transposed and PHYSICALLY connected in example sequence Section1PhaseA to Section2PhaseB to Section3PhaseC and vectorially adding these three voltages together yields zero under the above conditions. You may consult CIGRE Electra reports 28 from 1973 or #47 from 1976 or IEEE guide 575 etc.

 

[submonkey]

Hi HSKabel,

Thanks for your post - I agree with you.

The point I was trying to make, was that one should either:

(a) transpose the phases, and keep the sheaths "straight"

(i.e. One phase conductor spends 1/3rd length in far left position, 1/3rd length in middle position and 1/3rd length in right position)

- OR -

(b) transpose the sheaths, and keep the phases "straight"

(i.e. One complete sheath run spends 1/3rd length in far left position, 1/3rd length in middle position and 1/3rd length in right position)

But one cannot "transpose" both sheath and conductor in the same way.

Thanks,
Alan

Transpose the
Obviously either the sheaths, or the conductors must spend equal lengths

 

[davidbeach]

As jghrist stated, transposing the cables does not eliminate the need to cross bond the shields. A shield "circuit" needs to shield equal lengths of all three phases. The physical relationship of the phases has nothing to do with the currents that would be induced in the shields.

 

[cgrodzinski]

Sorry folks for "reheating" this a few months old thread. I was looking for information on the same subject and arrived here.
Most of what was written here is right. However, the cross-bonding system does not eliminate the circulating current entirely if it is used on the arrangement shown above (flat) because it is not symmetrical system. There would be some residual voltage at the end of a major section. Using trefoil arrangement would help.
The second note: the equal length doesn't eliminate the residual voltage either. Since the system impedances depend on the separation of phases the axial distance between cables must be equal through the entire major section. If you have horizontal directional drilling where the cable spacing is different the whole grounding system has to be carefully calculated. The same happens if you change cable arrangement at any point of a section.
Third note: if you use the single ended bonding system be aware that the ground continuity conductor (that must be used) is a parallel running conductor grounded at both ends. There would be some induced voltage and circulating current in this conductor. This in turn should be taken into account when calculating the system ampacity if the circulating current magnitude would increase the GCC temperature significantly.

In my opinion the sheath grounding system is difficult to calculate and most of the time neglected in the process of designing. Its modelling should be as careful as for ampacity calculation.
If any of my notes are not 100% correct please feel free to notify me or post your correction.
On the last note, please be careful when use IEEE575 existing issue.

Thanks