HV Cable Single Point Bonding within a substation

[ahandle]

G'day,

I need to run a 3.3kV cable from a 10MVA 33/3.3kV transformer to a prefabricated substation within the same switchyard. At 1750A, with derating factors due to the heating effects of parallel cables, it looks like I need 4 parallel runs of 630mm^2 1C cables (12 cables); presenting some installation difficulties.

Using single point bonding, the datasheets say I can instead do this with 3 parallel cables instead, however my understanding is that a voltage will be induced at the non-bonded end. I have seen numerous other posts on a similar subject, but these deal with long distances using cross bonding techniques. In my case however, the entire cable length is very short - only about 25m. What I would like to know is, what would the expected potential be? Will this be low enough that the cable can be safely run without making any typical installation modifications to a normally run cables?

The cable will be run undeground in conduit, coming up bottom-entry to a switchroom, 1.8m off the ground.


Many thanks in advance.

 

[ahandle]

Sorry, to clarify my query above, the bonding is relating to the screen. As these cables are within the switchyard compound, no metallic armouring is used, purely the heavy duty screen, rated for 10kA/1sec.

As another note, if the screen is only bonded at one end, that infers to me that there is no direct ground connection. Do I need to run another separate earth cable? If so, do I just use the one 120mm^2 earth cable or do I need one per parallel run? i.e. three cables?

 

[WhiteyWhitey]

All of the questions about bonding and induced voltage are related to dealing with voltages and circulating currents in the screen of the cable.

So if you have no metallic sheath (screen), then no problems with circulating currents or induced voltages.

As far as the earthing goes, I am no expert on earthing so a bit hesitant to offer any advice.

If your earthing grid is design to avoid any touch and step potentials at both ends of the cable then I cant see any reason why you need to run a separate earth cable with the 3.3kV's. You should check with your earthing designer though.

 

[WhiteyWhitey]

I forgot to mention I am not from the states and there may be requirements for an "Equipment bonding Lead"

check this thread

 

[7anoter4]

First of all you don't need a separate earthing cable [by the way: the shield of 3.3/1.9 kV cable is about 90 sqr.mm]. Second: I think the maximum build-up voltage will be, in any case, less than 25 V. This voltage depends upon current through the main conductor, the reciprocal position between cables [parallel rows, triangle etc.], the distance between conductors and, of course, the length of the cable. I think the best location of the earthing it is the far end from the source [all 12 cable shields bonded together].

 

[cgrodzinski]

I would suggest that you consult manufacturers of "cable bus". I do not know the details of your intended installation but installing them all together in one conduit does not seem to be a good idea to me. Note that the all cable metallic parts subject to mutual induction (no matter how long), either sheath or conductor or armour. If there is more than one conductor the cables should be installed in a "controlled" manner to have the mutual induction under control. Otherwise you will have to have a crystal ball to know which of them is overheated. Unless you know exactly distances between cables any sheath voltage calculation will be +/-50% (or less).

Even if you know cable separations calculating system parameters with more than one cable per phase is a tedious task.

Cheers,

 

[7anoter4]

I don't think that "The cable will be run underground in conduit" means that all 12 cables run in a single conduit but the cables run in a duct bank with-maximum- 12 conduits, usually P.V.C. made. So single core cable run in a single conduit each one or 3 cables per conduit [or duct].
Let's take General Cable 3.3/1.9 kV of 630 SQR.MM copper conductor .As indicated by Manufacturer ampacity of 3*single core cables in a p.v.c. duct [one single duct occupied] will be 551A per cable[1.2 Cm/W earth thermal resistance and 25oC earth temperature ].See:

http://www.generalcable.com.au/Australia/Products/energy/MV_xlpe/4.2.11.11.1.3.2.pdf

But, it is for shield both ends solid grounded[earthing ].If only one end of shield will be grounded then for 4 ducts filled with 3*1/c 630 3.3 kV cables[three phases per one duct] XLPE insulated ampacity will be 460 A.[ thermal resistance 1.2 Cm/W, earth temperature 25oC, distance between duct center lines 250 mm.]
The built-up voltage for 437.5 A per cable will be 3.2 V for 3 *1/c per duct[4 conduits] or 4.45 V for single core cable, each in separate p.v.c. conduit, 250mm apart[3*4=12 conduits].

 

[cgrodzinski]

It doesn't make any sense to me to build a duct with several conduits to run 3.3kV cables. The duct bank itself will cost many times more than the cables. In my opinion, for this purpose there are cable ducts that are designed by cable specialists that have available engineering calculation programs that help them in establishing the cable locations. I do not think that if the circuit load is 1000A using 4 cable rated 250A will be OK in every configuration. The mentioned table shows only one cable per phase. Am i right?

 

[Kiribanda]

ahandle,
As pointed out by WhiteyWhitey in his post,why cannot you use Aluminium (or non magnetic)armoured single core cables for your application to nullify all induced currents?

 

[cgrodzinski]

Kiribanda, as you noticed we are talking about the induced current. A magnetic field is always present when current flows through a conductor no matter what material is made of. If an another closed circuit conductor is placed in this field an electrical current will be induced when the magnetic flux through a surface bounded by the conductor changes. Either by moving the DC field or if it is time "dependent" as AC is. Sorry for this simplification.

 

[7anoter4]

Hi cgrodzinski,
First of all you are right: the General Cable table shows only one cable per phase ampacity.
Further, you are right also: one cable per phase ampacity is always more than ampacity of each one of four cables parallel in the same run.
You are right also: the duct bank is more expensive than simple underground buried cables.
But, some time you need to cross over a street or a highway and you have to prepare the place before. You can close then a half of the street permitting circulation on the other side.
There are other advantages of duct bank as, for instance, possibility of extracting of a damaged cable and put another in its place. Also you could provide a reserve duct for future expansion.
In order to calculate the ampacity of parallel cables in a duct bank one could follow the proceeding of IEC Standard 60287 or IEEE/ICEA S-135/P-46-426[J.H. Neher and M.H. McGrath AIEE Paper 57-660].See:

http://www.upcomillas.es/periodicas/Normas/IIT2902 iec60287-2-1{ed1.2}b.pdf

or:

http://neher-mcgrath.com/Heating_Calculations.pdf

In order to calculate the built-up voltage in the shield you may follow EPRI-EL5036 Power Plant Electrical ref. series VOL.4-WIRE AND CABLES.
There are some national standards as NEC [NFPA 70] indicating ampacity. For instance in
APPENDIX B single core cable of 1250 MCM [633sqr.mm] for only 9 duct bank, the ampacity of copper conductor insulated for 2000 V and 75oC maximum temperature it is 536 A.

 

[cgrodzinski]

Hi 7another4!
Thanks for the addresses on behalf of all participants of this thread.
However, the duct vs direct buried could be discussed. In the first post of this thread ahandle was connecting a transformer with switchgear. I would use rather cable bus than duct but he may have a different reason. Also using screened cable at this voltage may not be necessary. And again this is his choice.

Now, the duct bank is popular in NA. Sometimes there is no other way of doing it as you should not install direct buried circuits in couple of layers. This is for distribution circuits. Apart from this, the duct installation is not the best solution from the thermal point of view. "rho' of air in the conduits is ... 40 [K*m/W]. On the top of it there is a duct bank liner made of PVC with resistivity of 6.5. There are more "pros and cons" to this subject. In any case, I would not recommend using "standard rho" or standard soil temperature or backfill material without testing or circuit load factor taken from "the space". One can get a big surprise or loose a lot of money or both.

Going back to the extracting of damaged cable ... Yes, but you have to have quite a few meters of cable to sacrifice ... while mane repairs of direct buried circuits take just another joint.

Going again to the sheath voltage - everything can be calculated but it is a matter of complication and I doubt if there are many engineers that will be eager to do that for two cables per phase. You can check IEEE 575 or ELECTRA (forgot the numers ... there were two issues).