Cable Ampacity Calculations 4

[RIA123]

Gents
I am working on Cable Ampacity Calcs on EDSA software. Three single phase cables are going to a transformer and three out going feeders (9 cables-single phase cables) are placed in a concrete duct bank - all cables have concentric neutral wires for return current. The cables are designed to carry neutral current equal to 1/3rd of phase current.
The software can model the cable with concentric neutral wires - but there is no means to add this return current.
My problem is that how to take this neutral current in to account - do I have to add this neutral current into phase current and calculate total temperature rise?
- or does the software takes into account the neutral current once concentric neutrals are meodelled?
- or anything else?
- Can anybody give an idea?
Thanks

 

[cgrodzinski]

RIA123, have you consider grounding the concentric neutral wires? If yes, let us know how. The EDSA program follows the rules of IEC60287 so it must use the circulating current if any.

Thanks

 

[RIA123]

Thanks cgrodzinski - The cables are grounded at the substation - low resistance grounding.
The EDSA model can only show concentric neutral wires - there is no means to show whether the neutrals are grounded or not.

 

[cgrodzinski]

To be on the same wavelength - I am talking all the time about concentric neutral wires - NOT cable conductor.

I understand that you use "single end grounding" method - only at one end? Is this correct?? If it is, you do not have to worry about the cable concentric neutral (assuming that the concentric neutral wires have a jacket isolating them from the ground) as it will not carry any current.
If, like you said, there is no place to inform software what kind of grounding is I would call EDSA support and ask them how they modelled the algorithm for calculation. It is obvious that the grounding methods affect the cable ampacity if it is different than described above.

 

[jghrist]

Considering unbalanced current with a neutral return gets very complicated. Most likely, EDSA considers circulating current, induced from phase current (both ends grounded) but not unbalance load current. With unbalanced load current, some will flow through the earth, there will be less heat load from the adjacent cables because not all three phases are carrying maximum current (otherwise there would be no unbalance). Unless you have pure single phase circuits to consider, I would assume balanced loading.

 

[7anoter4]

I think the concentric conductor is like a shield made of round conductors. The shield losses are introduced as fraction of the main conductor losses as per IEC 60287-1 loss factor for sheath and screen [and it is not important where this shield is located-around individual core or around the assembly]:
The power loss in sheath or screen [lambda1] consists of losses caused by circulating currents [lambda'1] and eddy currents [lambda"1]. The eddy current in this case is negligible so only circulating current could be taken into consideration as: Ineutral ^2*Racneutral/ (3*Imainconductor^2*Racmain).
Racneutral could be Rdcneutral for 90oC if the insulation is XLPE and Racmain could be calculated from Rdcmain recalculated for 90oC and adding skin effect and proximity effect as per IEC60287.

 

[NAZ55]

I believe according to NEC ampacity is reduced if you are using more than three current carrying conductors (This applies to 3 phase circuits). Neutral is not considered as a current carrying conductor.

 

[NAZ55]

For single phase applications neutral is considered as a current carrying conductor and carries the same amount of current as the phase conductor.

 

[RIA123]

Thanks guys - for your valuable comments.

 

[WhiteyWhitey]

Just to add a note to anyone still watching.

I believe It is possible to model the effect of the circulating sheath currents without actually having a knowledge of the current magnitude.

If the temperature of the sheath is assumed to be maximum rated sheath temp (say 90Deg) then you could calculate its effect as part of a thermal circuit.

This would be effective for giving max ampacity of the cable. but may fall down in dynamic ratings and FEA studies at say half rated current.

Just a thought.

Andrew

 

[jghrist]

If the temperature of the sheath is assumed to be maximum rated sheath temp (say 90Deg) then you could calculate its effect as part of a thermal circuit.

If this were the case, then the ampacity would be zero. Any conductor current would increase the insulation temperature to more than the 90° shield temperature.

 

[WhiteyWhitey]

Hmmmm....

Good point. I dont know my modelling well enough to comment.
Anyone know where to get humble pie?

Andrew