Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IFRs on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Individually Screened Cable zero sequence impedance 2

Status
Not open for further replies.

acog

Electrical
Jul 30, 2010
31
Hi,

I have a supply substation with an earth grid, and a long cable (e.g. 10km) which supplies some equipment sitting on its own earth grid. I have a good understanding of the earth grid impedances of both the feeding substation and the remote equipment earth grid.

The cable which feeds the remote equipment is a three core XLPE compressed strand Aluminium cable where each core is screened by copper strand as per the image below.
cable.gif


My problem is that I need to calculate the percentage of current which enters the soil and the percentage of current which returns via the cable screen under an earth fault scenario at the remote equipment. The project is in the design phase so I can not test via injection.

I was intending to solve this problem by calculating the cable screen zero sequence self impedance (Zs) and mutual impedance (Zm) then using the following formula to work out the current in the remote earth grid.

Remote Ground Grid Current = Total Fault Current Available at Remote Source * (Zs - Zm)/(Zs + 3*Rg)

Where Rg = the remote earth grid resistance + the substation earth grid resistance.


However I can't find any method to work out the GMR (and hence the mutual impedance) for the cable screen because the screens are around the phase conductors. The only method I can find involves a cable with a screen which encircles all three cores, and I am not sure if this is applicable here.

The cable manufacturer has supplied 'Zero Sequence Resistance' and 'Zero Sequence Reactance', however I am not sure as to whether I can use these numbers for the cable screen impedance calculation.

Since the cable runs are long and there will be good inductive coupling of the cores and screens, I don't want to just use the DC screen resistance value.

Any help would be greatly appreciated.

Andy
 
Replies continue below

Recommended for you

Thanks for your reply MKFPE,

The document you linked is a good resource but does not explain how to calculate the zero sequence mutual impedance of the screen for a cable with screens around each core, which is where I am getting stuck.

Refer equation 9 and 10 from the linked document. I don't understand what to use for the GMR of the sheath (ri and ro).

Surely this is a simple problem for experienced earthing engineers who must calculate the fault current splits of cable screens / ground regularly?


Andy
 
Andy,

Sorry I read your post quickly, not thoroughly.

Kersting's book "Distribution System Modeling and Analysis" covers calculation of impedances of concentric neutral cable. Cooper's "Distribution-System Protection Manual" does as well.

A Cooper sales engineer gave me a free copy of their book years ago, I don't know if they still do that. Kersting's book is one of the best investments any power engineer can make. The circuit matrix representations can be set up directly in Matlab or Freemat or Octave, or similar, essentially as outlined in the book, and by changing parameters for your specific problem, "job done".

I can't provide copies of the applicable pages due to copyright, but I hope this provides some direction: Calculate impedances by procedures given in these references, do sanity check by comparison with published tabular values such as in the Kerite Engineering Catalog (on-line), follow method in the NZCCPTS document to complete the calculation.

BTW, this is not a calculation that I recall having had to make, I am primarily a substation engineer, my power cable shields are open-circuited at the remote end and a separate neutral conductor is used. But the above is how I would proceed.
 
The following text originates from the BICC Electric Cables Handbook, Third Edition, pages 175-177. Refer to the calculations for SL type cable.

Zero sequence resistance
The zero sequence resistance of the conductors per phase is the a.c. resistance of one conductor at ambient temperature (normally 20°C) without the increase for proximity effect described in chapter 2. The increase for skin effect is included. To this value is
added the following as appropriate:
for 3-core cables:three times the resistance of the metallic covering;
for single-core cables:the resistance of the metallic covering;
for SL cables:
the resistance of one metallic sheath in parallel with three times the resistance of the armour.

The metallic covering for 3-core and single-core cables may be a metal sheath,armour or a copper wire and/or tape screen, whichever is the metallic layer which will carry earth fault current. If the cable has a metallic sheath and armour, the resistance of the metallic covering is the resistance at ambient temperature of the two components in parallel.
Some 3-core cables may contain interstitial conductors or copper wire screens on each core to carry earth fault currents. In general, where there is such a component for each core, it is the resistance of one of them which is considered. When the earthing component is a collective one, i.e. a common covering for all three cores, its resistance is multiplied by three. Where both types of earth current carrying components are present, the example given for the SL cable (three sheaths but only one armour) is followed. In effect the calculation is equivalent to determining the resistance of all the earth paths in parallel and multiplying by three for the resistance per phase.
The resistances are calculated from the cross-sectional areas and resistivities of the metallic layers. For a sheath the cross-sectional area can be calculated from the iameter and thickness. For wires, the resistance of a single wire may be divided by
the number of wires and a correction made for their lay length as applied. When wires re applied helically or in wave form, their individual lengths exceed the length of the cable by an amount depending upon the ratio of their lay length to their pitch diameter.
For example, armour wires applied with a lay of eight times their pitch diameter have a length approximately 7.4% greater than the length of the cable. Technical Data Applicable to Cable Planning and Usage 177
Zero sequence reactance
For single-core cables, the zero sequence reactance of the cable per phase can be calculated from the equation
X0 = 2 pi f × 10 -3 [0.21oge(D/d) + K] (Q/km) (10.6)
where X0 = zero sequence reactance
f = frequency (Hz)
D = mean diameter of metallic covering (mm)
d = conductor diameter (mm)
K = a constant depending on the conductor construction as in chapter 2,
equation (2.2)
The similarity between this and the combination of the equations for inductance and reactance in chapter 2 is evident. The value D, which amounts to twice the spacing from the conductor axis to the metallic covering, replaces the 2S where S is the axial spacing
between conductors in normal operation.
For a frequency of 50 Hz the equation simplifies to
X0 -- 0.31410.46 lOglo(D/d) + K] (Ohm/km) (10.7)
For a 3-core cable the equation is
Xo = 0.4341oglo(D/GMD) (Ohm/km) (10.8)
where GMD is the geometric mean diameter of the conductors in the laid-up cable. Conventionally, the value of GMD is taken as 0.75 of the diameter of the circle which circumscribes the conductors in the laid-up cores, assuming the conductors to be circular.
The equation for the 3-core cable is analogous with that for single-core cable, the value K now not being relevant and the 0.434 including multiplication by 3 for reactance per phase.


Regards
Marmite
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor