Circulating Ground Current on Cables

[vtpower]

One more question on a totally seperate issue. At one of our substations, we have 15KV cables coming off the low side of a 34.5 / 12.47 KV transformer. This feeds our low side bus which feeds the four feeders. The cables are Jacketed Concentric Neutral cables and there are 3 - 500MCM copper cables per phase. All three of the same phase are in their own pipe, so there are three pipes, each pipe has a seperate phase. During an IR check, we noticed one of the concentric neutrals where it ties into the ground grid was hot. We amp probed it and found 300Amps flowing on the concentric neutral??? We checked the other three phases and found the same thing. This neutral is tied into the ground grid and all three neutrals are tied together. The only though was induced voltage on the neutral is causing a current flow but it is a very short run and this doesn't seem to make sense. Again, any insight anyone might have would be very appreciated. We also checked amperage directly on the Xo bushing of the transformer and found next to nothing as expected because of the balanced system.

Thanks.

 

[waross]

The phase current is inducing a voltage in the concentric neutrals. When you grounded one end of the concentric neutrals and connected the other ends you provided a low impedance closed circuit for the induced currents.
I have experienced badly overheated cables to the point of causing severe heat accelerated corrosion at a 400 amp 208 volt switch. The problem was caused by running "A" phase and "B" phase in one steel pipe and running "C" phase and the neutral in a second steel pipe. The load never exceeded 200 amps but the 400 amp switch had been destroyed by heat accelerated corrosion twice before I corrected the arrangement. This is not the same effect exactly as your problem but it is related.
The point is that distance is unimportant. With a short length of cable the induced voltage is less but the loop impedance is correspondingly less.
Your induced voltages and circulating currents will increase as the line current increases.
respectfully
If your system requires a neutral use a neutral cable instead of the concentric neutrals.
I suggest grounding the concentric neutrals at the supply end and then keeping them insulated from ground and each other. Insulate the concentric neutrals at the load end and do not connect them to anything. Depending on your layout you may consider grounding at the load end and insulating at the supply end. The neutral will become a ground conductor in the event that a cable fails so make your connections to ground well.
If your pipes are ferrous you will probably be experiencing heating of the pipes also.
respectfully

 

[apowerengr]

I agree with waross - ground the supply end and isolate/insulate the load end. You can then run a dedicated cable from the Xo bushing down the side of the transfomrer to the ground grid and use the ground grid as your neutral. Bond your feeder neutrals to the ground grid. This should take care of your heating/loading issue. Note, this opinion is based on the following ASSUMPTIONS: ground grid is adequately sized for the fault and load currents, you can get access to the Xo bushing and connect an adequately sized neutral/grounding conductor, your transformer protection can handle this connection without bypassing other protection, your feeder neutrals can be bonded to the ground grid, and that your operating practices accept the use of the ground grid as a neutral as some utilities don't allow this.

 

[davidbeach]

If your "pipes" are made of ferrous metal, or have any ferrous metal components or mounting brackets or anything of the sort you will have induction heating problems. How that might affect current induced onto the neutral, I don't know.

 

[waross]

Do you know that all three of one phase in one pipe is contrary to accepted practice?
Each pipe should have "A" phase, "B" phase, and "C" phase.
This will give less reactive voltage drops and avoid the ferrous heating that davidbeach and I have mentioned.
respectfully

 

[jghrist]

I would also run a 600 volt neutral cable with each ABC phase grouping and use LC tape or wire shield power type cable instead of concentric neutral URD cable. The higher resistance of the tape or smaller wire shield will reduce circulating currents. I would ground both ends unless calculations are made to ensure that induced voltage is not excessive with one end grounded.

 

[vtpower]

Thanks for the respones.

I am aware that this is not common practice to run similary phases in the same pipe. The pipes are all non-ferrous and do not have any ferrous material, i.e. condulators on them. I think we will end up having to do as Waross mentioned and isolate one end of the neutral from ground after we check the induced voltage capability on the neutral. Thanks for the info.