connection of PT to a ground that's not there

[mshields]

Have recently encountered a 35kV switch with a CPT that has a 19.9kV primary with 120/240V secondary. The problem is that the incoming 35kV has 3 phase conductors and no other conductor of any kind. i.e. no ground, no neutral, no grounded service conductor, whatever you want to call it. There is no path back to the source for any single phase loads. Now, there is a ground for part of the run. This is a college campus with 35kV primary metered service. The line coming into the campus has no ground as I say. This is an overhead service but once it goes underground they DID put in a ground. So, there is no complete path back to the source might be a more apt way to put it.

Now, the CPT is working therefore the current is finding a way back first via the ground in the underground segment which tie into the substations ground grid and presumably find it's way back via a variety of paths in accordance with ohm's law.

I'm wondering just how dangerous this is? What do you think?

 

[JFletcher]

Pretty much our entire neighborhood is wired that way; 34,500kV 3 phase overhead, with pole mounted 120/240V transformers fed from one phase of 19920V. The "source" is the xfmr, not the service. I'm probably not understanding your concerns...

 

[K8MHZ]

Pretty much our entire neighborhood is wired that way; 34,500kV 3 phase overhead, with pole mounted 120/240V transformers fed from one phase of 19920V. The "source" is the xfmr, not the service. I'm probably not understanding your concerns...

I'll bet if you look, there is a 'static line' bonded to the xfmer can and grounded at the base of the pole.

 

[mshields]

clarification

clarification

The secondary is not the concern. I'm concerned about the primary. When you draw current from the secondary, current flows on the primary. How does it get back to it's source as it would have to do in order for current to flow?

In my case, it is apparently finding a way. But that way is not via a conductor or metal parts that are supposed to carry current. I'm wondering what safety concerns exist in such a scenario?

 

[Ingenieur]

The switch enclosure is not grounded?
if not by oh conductor it may have a ground mat
I assume since the pt is functioning the system is grounded (feed is wye)
otherwise a phase to frame fault would do nothing until a second ph faulted

either way the primary current is minute
if sec draws 10 A of 120 vac control power the primary flows 120/34500 x 10 ~ 35 mA

 

[Bugman1400]

In residential application, I would expect a dist class single phase xfmr to be tied to a neutral somewhere and ground. However, in a substation, a single phase CPT is always tied to ground via the casing, to the structure, and down to the ground mat pigtail. A phase-to-phase CPT could have also been used but, grounding of the case and secondary will still be required.

A picture would be great.

 

[mshields]

in answer to your questions

in answer to your questions

Everything is grounded appropriately, there is an extensive ground grid and everything is interconnected.

The only concern is that between the campus and the utilities substation, several miles away, there is no 4th conductor of any kind on the pole line. To complete the circuit, somehow that 30mA needs to get back to it's source. Obviously it is finding a way. Is 35kV sufficient to have it pass literally through the ground? Maybe it catches a ride on interconnected steel water pipes? who knows?

I see your point that there are only milliamps passing through that circuit however it is completed. But 30mA is decent amount if you become part of the path back to the source. That is, I will admit hard to imagine if everything is grounded and at the same potential.

Bottom line, it is not an inappropriate situation to have no ground conductor or neutral on the pole line carrying return current back to the source and I'm struggling to quantify what my practical concern ought to be if at all.

 

[big john]

Sounds like the CPT is running on earth-return. What is the other side of the transformer actually connected to? The ground mat? What is the conductor type of the grounded leg?

I'm not a fan. I can picture a scenario where someone opens what they believe to be an electrode conductor only to find themselves on the business end of 20kV.

 

[don_resqcapt19]

Around here, they use a common neutral for the primary and the secondary, so it looks like there in no path for single phase primary current but there really is.

 

[Ingenieur]

The earth path between the station bed and switch mat is much lower than you imagine
R = p L/A
and A ~ L^2
I've measured small station beds using a shallow single rod at 200' distance at <2 Ohm
the further out the lower
it has to return to the xfmr X0 (zero seq reactance) KVL and KCL, loop must be satisfied

look at it like this
V gnd at sub and switch is ~ 0, so drop is basically 0
V = iR
if V is 1 volt ad i is 0.035 mA R is 28 Ohm
it is much less
the ct |Z| is 34500/1.732 / 0.035 = 570 kOhm
the earth path Z is inconsequential

 

[GoldDigger]

The only concern is that between the campus and the utilities substation, several miles away, there is no 4th conductor of any kind on the pole line. To complete the circuit, somehow that 30mA needs to get back to it's source. Obviously it is finding a way. Is 35kV sufficient to have it pass literally through the ground? Maybe it catches a ride on interconnected steel water pipes? who knows?

If all of the transformers at the substation have their primaries connected only line to line, then there would not be any current that needs a neutral to return to the source.
If there are some line to "neutral" loads at the substation, it is practical at that voltage to use an earth return, but AFAIK that is not done in the US.

The issue is not whether 35kV is enough to pass through the ground, it is what the voltage drop at the earth electrodes at each end would be at the current being sent through the earth. An earth electrode (grid) specifically made for that purpose could have a contact resistance on the order of an ohm or less. If the earth current is 100A that would only be a 200V drop against the 35kV of the circuit, so the percentage VD can be very small. On the other hand, the 100V drop across the earth near the electrodes could cause dangerous touch or step potentials unless they are protected by an equipotential grid within the substation.

As far at the PT is concerned, a voltage on its primary of 19kV on a 35kV three phase circuit does suggest that it is in fact wired line to earth. As long as none of the load current is traveling through the earth, AND the source is actually a grounded wye or a delta with a zigzag transformer for ground reference, then the small current trough the PT could use an earth return with no problems at all.

 

[Smart $]

... the incoming 35kV has 3 phase conductors ...

... To complete the circuit, somehow that 30mA needs to get back to it's source. ...

How much has to find its way back to the substation depends on the phase to neutral (or earth) imbalance locally. With 3 primary phase conductors being run, I'm quite certain there are other nearby transformers, and quite possibly line to neutral (or earth) ones at that.

The question is whether there's a local zig-zag for earth reference.

 

[gar]

160921-2015 EDT

I don't understand the original post and the relation of some of the posts to that question.

My interpertation of the original post is that mshields has what appears to him to be an ungrounded 35 kV delta source at a location where he has a 20 kV potential transformer connected from one hot line to earth (but earth was not mentioned, the second side of the potential transformer had to be connected somewhere, and the primary was rated at 20 kV, line to neutral voltage is likely 20 kV at the substation transformer)..

His question seems to be how can this work?

One assumption is that the 35 kV source is really a wye secondary with its neutral point connnected to earth at the wye transformer location. Othetr means could be used to created a ground reference.

The primary current into a potential transfomer with a 20 kV primary is going to be about 120/20,000 of the current required by the voltmeter load on the potential transformer. To this add some core losses, probably greater than a meter load. Good quality AC moving coil voltmeters can be in the range of 1 mA. Thus, the meter component at the primary could be 120/20,000,000 ampere, or 60 microamperes.

If our meter accuracy was 1 %, then 1 % corresonds to 350 V at the PT primary.

Once you get beyond near field voltage drop in the earth the earth resistance appears to be a very low resistance.

AC voltage drop measurements over a 12 ft distance in my backyard are usually less than 0.2 V RMS. This is between two screwdrivers poked into the ground.

I believe that any voltage drop thru the earth for the original post potential transformer is small by comparison with voltage drops in the earth resulting from other current sources. These other sources would be the primary cause of voltage errors seen at the potential transformer output terminals.

.

 

[Ingenieur]

He called it a cpt so I assumed power for controls/relaying/etc
he said other ckts from the same source have grounds
he did not mention a zig-zag
he said it works so must be a complete ckt
he said it has a ground bed and seemed to infer it is bonded to it

I guess it could be a delta with zig-zag to derive a ground
I usually see that on a delta sec to impose a ground |Z|

we are all guessing

 

[GoldDigger]

He called it a cpt so I assumed power for controls/relaying/etc

The thread title said PT (i.e. for measurement only) instead of CPT, and I think some members were responding to that instead of the CPT term used in the body of the post.
You are right that we do not know enough by quite a lot.

 

[big john]

While I imagine this is system is wye grounded somewhere, even if it weren't, I wouldn't be atall surprised if a there was a few hundred mA of current flow just from leakage and coupling.

No different than how ground-detection lamps will illuminate when eathed on a low-voltage ungrounded system.