PARALLEL(ING) HV UNDERGROUND CABLES 4

[SphincterBoy]

I have a situation where I need a jacketed, concentric neutral, underground cable of size 1500 MCM (minimum) per phase.

Looking back on past experience with failed HV jacketed underground cables, it is *highly* desirable to install a cable whose size does not exceed 500 MCM per phase, in order to ease pulling of the cable in PVC ducts. In other words, 1500 MCM cable is just too big to pull out and replace easily.

Would it be practical to parallel more than one 500 MCM cable, to create an equivelent 1500 MCM cable?

Has anyone ever done this before?

My chief concern is the cable characteristics between each conductor, although slight, may be enough to cause significant differences in current as the loading is increased. The only thing I can think of is to place CT's on each conductor, and meter the current flow for each.

 

[dpc]

There should be no major problem paralleling cables. The main concern is to make each cable as close to the same length and configuration as possible.

I've seen many installations with two or three 15 kV conductors per phase. I've have not seen anyone install individual CTs, although that might not be a bad idea. Probably thermography of the cables and terminations periodically would be a good idea as well.

Of course, for low voltage feeders, it is common to run many conductors per phase. I've tested quite a few of these when taking field measurements and have found that the current generally divides pretty evenly, unless there is a loose connection or termination.

 

[BJC]

As dpc said it's done all the time. The longer the cable the more they even out.
Try using Xcel and playing around with the change in temperature vs resistance. It doesn't mean much practicaly but it's interesting.
Your CTs would have to go inside the concentric neutral of the cable may be hard to do.
You could consider a thermocoulple or RTD in the duct.
Two 500s should do as well as a single 1500 MCM ( assuming you could buy shuch an animal ).
How are you going to protect the cables? How long lis the run?

 

[busbar]

Side note — When using multiple MV conductors per phase, clearance requirements are a bit different, exampli gratia,

http://www.raychem.com/products/epd/rrg/98.pdf

 

[SphincterBoy]

Thank you for the quick response!

I feel a little easier now.

The FLIR scan is a good idea. Loose/poor terminations have really plagued us over the years. We are also going to do a 2 min. "midnight" photo for corona.

As far as the type of cable is concerned, I am planning Kerite (the best). The total circuit length will be about 1500-2000 feet, inside PVC duct. I wanted to 'break' the run into two sections, and place an enclosed junction, but have since backed away from this idea, since the junction is located in a area where there are no roads.

We always use MOV's on each end of any underground cable, and we've found they really work. There will be three-phase isolation switches at each end (S&C Electric) as well.

I like the idea of using temperature detection inside the duct in addition to metering each conductor.

To Busbar: Thanks for great info. on the spacing!

 

[busbar]

spink — As for current sharing in parallel cables, what about floating some medium-sized metering-class CTs—slid over each cable behind the terminations? You could even connect a suitably long SIS loop to later check for sharing with a low-range ‘clothespin’ clipon CT, and abandon them in place after initial checks. If the CTs were secured with nylon ties below the cable-termination bases, secondary grounding probably wouldn’t be of concern for the short term.

Larger relay-class CTs probably wouldn’t be necessary for verifying steady-state current division.

Kerite Rules!!

 

[busbar]

For ANSI-based installations, NEC 310-4 is good baseline practice. With runs of that length, current sharing between of per-phase parallel conductors should not be a serious problem. Now OTOH, if the span was ten feet, it's a bit tougher task. Oh, and on the Kerite, hydraulic-compressed 2- [or 4-] hole ½-inch copper lugs are especially nice frosting on the cake.

 

[SphincterBoy]

Thanks for all the great info Busbar!

 

[dpc]

If you're not familiar with Calvert Cable Bus, you might want to check this out:

http://www.calvertbus.com/cablebus.html

May not apply to your situation, but it is a useful product in many circumstances.

Also, I have done designs using Trenwa trench systems with multiple metal-clad (Okonite CLX)cables per phase in the trench using a special thermal sand backfill. Spacers are used to keep an even separation of the cables. This allows direct access to the cables if necessary since the Trenwa is flush with grade. We were able to do 3000A at 13.8 kV using 6 750kcmil conductors per phase.

Another issue that I'm sure you're aware of is the necessity to de-rate the cable ampacity if you are running multiple conduits in the same duct bank.

 

[busbar]

spink ‼ Cables I can handel. Them pretective raley's just slay me, through.

 

[jwerthman]

I agree with the earlier comments regarding equal current sharing between the conductors. For the circuit lenth you are talking about it shouldn't be any problem.

One trick that can be used to force equal sharing in a situation like this is to put a CT on each conductor and connect all of the CT secondaries of each like phase in series. Because of the series connection of the secondaries, the secondary current will be equal in every CT, forcing the primary current to be nearly equal as well. I have not seen this done on cable circuits (probably because it's not necessary), but I have seen it done on paralleled circuit breakers and it works quite well.

 

[jbartos]

Suggestion: The use of subconductors is often preferred to the larger sizes of fewer subconductors or perhaps one conductor. It is necessary to take into consideration the physical layout of the subconductor run. If these subconductors form a tight bundle, then the electrical center of the bundle is negligibly off the geometrical center. However, if you run the subconductors in a flat layout the electrical center is somewhat off from the geometric center. This impacts the line reactance and impedance. Often, textbooks and problem solvers include these types of applications and examples.

 

[johnspark]

Comments: I too have seen CTs on paralleled conductors within CB busbar ducting to balance out current in the paralleled conductors. But for the 1500-2000 feet distance the, emf required to balance the currents means the balancing CTs will have to be very big (educated guess).

Furthermore, if you are on good terms with the cable manufacturer, they will probably give you the cable layout for the installation of the three phases to minimise mutual coupling effects that derate cable current carrying capacitiy

 

[SphincterBoy]

Thanks for all the great information fellas! This Eng-Tips forum is fantastic!

 

[SphincterBoy]

Another question: Does anyone feel I should use a metallic (tape) shielded cable? I understand the metallic tape shield is to primarily keep moisture out of the primary insulation, and for a very low impedance.

Yet another question: I have to route the conductors underneath a RR track. I am tentatively planning on using sch. 80 PVC at a fairly appreciable depth underneath the track, minimize movement and/or damage to the conduit. Should I use a concrete-incased ductbank?

 

[dpc]

I'm not sure where you're located, but if this RR track belongs to a railroad company, you'll need a crossing permit. And they will probably tell you how to put it in.

Are you planning to excavate, or are you going to bore underneath the track?

 

[SphincterBoy]

We plan on boring.

 

[dpc]

The last time we did this, we put the conduit inside a large cast iron pipe to allow for future replacement/additions. I believe this was at the request of the railroad. We did not concrete encase anything.

Anyway, dealing with railroad real estate people could be the most difficult part of this project.

 

[SphincterBoy]

DPC, thanks for warning about the railroad. Did the iron pipe lower the ampacity of the conductor? Also, iron rusts.....yuck.

Many thanks for all your comments!

 

[ausphil]

Just another side of operation of the system is ease of fault finding. If you are going to run parallel cables, just make sure that in the future, you don't decide to join them together along the run or tees off them (for what ever the reason at the time). We have a couple of cables that have a multitude of single cable and parallel cable sections, then just to complicate things, there is tee-offs in both the single and parallel sections (yes parallel tees off the parallel sections) which have arisen over many years of system alterations and growth.

Whilst in operation, the system is fine, however, when a fault occurs, it can be near impossible to find (eg o/c faults in one leg of the parallel section) and there can be some pretty serious delta-star resistance conversions when trying to work out bridge locations. I know that you'll never let it get to this stage, but just keep in mind what your successors may do.