Objectionable Current

[benray]

I have been in the electrical industry for 50 years. I am now engaged in performing commission inspections and related tests of new building electrical systems.
I am finding current on all equipment ground conductors to power transformers in the facilities.
This is due to the X-O of the secondary being effectively(earthed) at a point other than at the service ground electrode location.
I have been insisting the X-O ground conductor be routed back to the main service ground point.
The latest edition of the Soares Grounding Book,illustrates the many grounding methods that create two or more ground paths and points.

I would like some of the intellects on this forum to give me their opinion of Chapter 12, 8th Edition, of the Soares Book. This publication is now sponsored by the IAEI.

I feel this book is incorrect and creating many inferior designs of ground systems, with objectionable current flow from interior and exterior systems.

On another issue;

Generator separately derived system in same publication.

I can not see where the utility neutral is being disconnected by the transfer switch, when it is shunted by the ground conductor.

Need all the expert comments I can get. Thanks; Bennie

 

[dpc]

I don't have the Soares book you reference, but if you are talking about a separately derived system, such as a transformer, the NEC requires that a grounding electrode be installed as close as practical to the transformer. I have never seen the grounding electrode conductor run back to the service ground. But the grounding electrode for the separately-derived system must be bonded to the overall grounding system.

I'm not sure what equipment ground you are measuring current on. Is this on the primary of the transformer?

Maybe I'm not clear on your question...

 

[benray]

Thanks for the reply. The equipment ground conductors, with current flow, are from the service panel to the transformer primary, installed with the supply conductors.

 

[benray]

I mis-stated the last message. The equipment ground conductors with objectionable current are from the service neutral/ground bus to the X-O of the transformer secondary.
This X-O is connected to effective grounded building steel at closest point.

 

[dpc]

OK, I think I get the picture. You're referring to the green wire running with the transformer primary feeder?

You may have some stray ground current in this wire, but it would not be objectionable unless the current belonged in the neutral (grounded) conductor. I'm not sure where you are measuring the current, but you seem to be assuming it is coming from the neutral point of the transformer secondary. I'm not sure this is the case.

On the secondary side of the transformer, assuming a wye connection, the neutral/ground current is trying to get back to the transformer neutral. If the secondary system is wired correctly, the only path is via the white neutral conductor, unless there is a ground fault or incorrectly wired device.

If you unground the transformer neutral and the ground current goes away, then you have a problem in the downstream secondary wiring.

But there are other reasons that some current may be flowing in the equipment grounding conductor (green wire) that are not related to the secondary grounding. Because the green wire is tied to ground at every box, panel, etc, current can flow due to small ground potential differences. This is true even when dealing with two ground rods - there can be measurable current flow through the earth. This is why special waveforms are used for doing ground resistivity measurements.

If I've answered the wrong question, let me know.

 

[benray]

I am seeing current from the load imbalance and from exterior sources on the equipment ground conductors. I read high EMF on these conductors.

 

[jghrist]

Is the load unbalance on the secondary of the transformer? What is the transformer connection? If the transformer is delta-wye, then there is no connection between the primary neutral and the secondary neutral. Load unbalance on the secondary side will flow in the secondary neutral back to the transformer. On the primary side, the unbalance will be in the phases only, not in the primary neutral or ground.

The transformer is a separately-derived source and the secondary neutral should be grounded at the transformer or before the first overcurrent device or system disconnecting means. See NEC 250.30. The primary neutral, if it is brought to the transformer, should not be grounded at the transformer.

I don't have a copy of the Soares book, but generally, you don't want to ground the neutral in more than one place or you may get objectionable ground currents. See NEC 250.6. "The grounding of electrical systems, ... shall be installed and arranged in a manner that will prevent objectionable current over the grounding conductors or grounding paths."

 

[benray]

Gentlemen: You don't know how much I value this information. I only hope it is not wasted on me, at the ending of my career. I only wish the internet was active 50 years ago. I,m sure others will read this thread and benefit from the years of searching my brain.

I agree a transformer is a separate source, but only when the neutral is not connected to the supply neutral. An equipment ground conductor from the neutral/ground buss of the supply to the derived neutral, excludes the transformer from the definition of a separately derived system.

The original description of a separately derived system was; Alternating Current Systems without Exterior Connections. The designation was only for calling attention for purposes of secondary grounding if required to be grounded.

 

[benray]

I have been involved in research and design of ground systems at some of the most unique locations in the world.
The conditions covered both extremes of earth impedance.

The coral islands in the Pacific constituded a variable resistor. The impedance changed in relation to the ocean tide.

The sand of the Sahara Desert in Libya, was silca, which offered a very high impedance.

The ice cap at the South Pole, is a near perfect insulator.
Earth connection was impossible.

My interest in ground systems, is based on preventing current flow where not necessary.

The earth is a giant and complex voltage divider. Connection at any two points will create current flow, in systems other than the supply source.

Current may return to its own source, but it can flow in many systems on its way home.

The procedures for grounding transformers indicated in the NEC, create two points of earth contact.

 

[jghrist]

benray, I'm not sure I follow your point about the transformer not being a separately derived source. Generally, a delta-grd wye transformer is considered a separately derived source. The equipment grounding conductor in the primary circuit is naturally connected to ground. So when the derived neutral is connected to the ground bus, it gets connected to the primary equipment grounding conductor. If this excludes the system from being separately derived, when would you ever have a separately derived system?

The NEC definition of a separately derived system says "...no direct electrical connection, including a solidly connected grounded circuit conductor, to supply conductors originating in another system."

The equipment grounding conductor is not a grounded circuit conductor.

Unbalance current in the secondary circuit of a delta-wye transformer will not flow back to the source of the primary system. This current comes from the secondary of the transformer. If you look at the zero-sequence equivalent circuit, there is an open on the delta side of a delta-wye transformer, keeping any zero-sequence current out of the primary side.

 

[benray]

Thanks for responding.

The supply conductors originating in another system are four wire secondary to a delta primary. The ground/neutral of the supply, connects to the X-O secondary of the user transformer.

The MGN is complete to the transformer secondary. This is a MGN system, not a separately derived system.

The phrase "separately derived system" is a technical procedure to make known there is no ground connection to the utility ground, and the secondary may or may not be connected to ground at the user facility.

End user load current is not the problem. The multiple ground points are the problem, when each transformer is connected to ground at a different location. This is the original reason for a transformer being described as a separately derived system.

The only common feature, with the list of equipment designated separately derived systems, is the ground conductor not being electrically connected.

A separately derived system (transformer)is when the primary is supplied by a 3 wire delta or a floating wye secondary from the other system.

A site generator is a separately derived system when the utility MGN is not electrically connected to the user service. This is to insure there is no connection to the user service during a power outage.

The unknown status of the ground/neutral during a utility transformer failure, is one reason for the isolation.

 

[benray]

I wrote the above to quick. Sorry if it is not clear. I wish there was an edit feature on this forum.

I would like to know the technical facts that determines an equipment ground conductor to not be a circuit conductor. Any conductor that carries current or is capable of carrying current is a circuit conductor. NEC 200.3.

Is there a document that substantiates the contradiction of the definition?

 

[busbar]

As stated, for the plain-vanilla drytype 480∆-208Y/120V 3–to-750kVA transformer have two grounding tasks, one for primary windings and enclosure and one for secondary windings. Both are intended mainly to provide rapid clearing of associated overcurrent devices and limit potential difference between conductive surfaces.

For these voltages, one case where the grounded-circuit conductor would have to be routed from the source would be for 480Y/277-208Y/120V autotransformers. I have not seen any of these recently, but Square-D used to {mid eighties} sell them. A variation with operation of these was that they freely passed triplen-harmonic currents.

For conventional isolating transformers, low-impedance paths between both circuits implies unavoidably, not-necessarily-undesirable parallel paths except in the relatively rare case where transformer-secondary bond is to a misguided, indeed gothic, remote-flowerbed ‘isolated’ electrode. One of my favorites had a remote electrode tied with dedicated, parallel 4/0AWG welding cables to 100kVA “Topaz-brand” transformer XO-bus stubs.


Here was a longstanding {since-modernized} university ‘research-lab’ standard…
· For transformer primary: metal raceway with bare equipment ground
· For transformer secondary: ‘jumper’ from XO term to robust building-perimeter ground ring—via 4AWG Cu THW white, {in 1-inch GRC} joined only at outdoor ground well.
Building ground-ring standard boilerplate was 4/0AWG 7-strand SDBC with minimum four 10-one-inch copper-clad rod. This was seldom not directly and uniformly tied to all AC-power grounded and grounding conductors, air terminals, building surge protection and virtually all research/particle-beam systems and enclosures.

Point is, each separate characteristic insulation failure mode needs to be allowed for, and high- versus low-side winding interaction during this does not generally cause any undesirable effects with respect to ground currents.

 

[benray]

busbar: I like your approach.I still question the equipment ground conductor with the transformer supply, not being regarded as a circuit conductor.

The origin of the terminology "Separately Derived System" section 250-26 first appeared in the NEC in 1968. In the 1965 Edition title was; "Isolated Systems".

Until the 1959 edition, the title was "Alternating Systems Without External Connections", Section 2514.

"Separately Derived Systems" was adopted as a term for including all power production equipment, not electrically connected to the servicing utility system. Electrically connected as defined by Section 200.3.

Section 200.3 Connection to Grounded System, specifies a grounded premises system, when connected to a grounded supply, is an electrical connection to the utility MGN system. An complete electrical connection is a circuit.

The grounded premises wiring, when connected to an ungrounded supply, has no direct electrical connection. The connection is by magnetic coupling, therefore it is a separately derived system by the NEC definition.

All schematics from the 1940 Handbooks, illustrate the utility lines as being ungrounded when the transformer is regarded an "Alternating current System without External Connections".

 

[benray]

Another issue; The illustrative technical publications, describing a separately derived system, specify that all transformers meet the definition excluding an auto-transformer. The reason is the supply neutral is connected to the auto-transformer.

How about a grounded wye/grounded wye, configuration? The supply neutral must be connected to both winding star points.

How about an open wye/open delta? The MGN ground/neutral conductor is electrically connected to both sources and carries load current.

Open wye/open wye, same situation.


All auto-transformers do not require a ground/neutral connection. Example 208 to 240 volt single phase.

I am sorry, I am missing something in the translation.
This is why I need the intellects on this forum to set me straight.

Thanks to all

 

[benray]

Another reason I would like to run this issue by the professional engineers on this, and other forums is;

The schematic of a transfer switch, in the Soares Grounding Publication, is identified as a separately derived system, due to the neutral being switched.

Please examine the equipment ground conductor routing and connection points on this drawing.

My High School basic electricity course taught me the equipment ground conductor is in parallel with the neutral conductor. The switch position has no effect on the electrical connection and continuity of the neutral.The switch is effectively jumpered out of the circuit.

This same logic is in the determination that an equipment ground conductor is not a circuit conductor, when it electrically connects the transformer neutrals.

Does the green insulation prevent current flow? I don't think so.

 

[jghrist]

The NEC distinguishes between a grounding conductor and a grounded conductor, with the grounded conductor clearly defined as a circuit conductor by NEC Article 100. The grounded circuit conductor carries unbalanced load current. The grounding conductor is not meant to carry load current, only to provide a low impedance path for ground faults for fast clearing and to provide low ground potentials for safety during a fault.

See the comments in the NEC 2002 Handbook under 250.4(B)(4). "Grounding can be divided into two areas: system grounding and equipment grounding. These two areas are kept separate from each other except at the point where they receive their source of power, such as at the service equipment or at a separately derived system."

Utilities (not subject to the NEC) normally use a multigrounded primary neutral instead of a separate grounding conductor. The utility service conductors also usually use the neutral as a grounding conductor and it is grounded both at the transformer (and connected to the primary neutral) and at the service. Because of this, stray currents can flow in grounding conductors, but these normally follow the phase currents because of inductive coupling and will not normally affect premises equipment grounding conductors.

The equipment grounding conductors of circuits connected to separately derived systems do not have to be switched like the grounded neutral conductors because they do not carry load current.

The secondary of a grd wye - grd wye transformer is not a separately derived system. The unbalanced load current in the secondary is transferred to the neutral of the primary. There is a connection between systems in the zero-sequence equivalent circuit.

All grounding conductors in a premises should be electrically connected together, even in the unfortunately named isolated ground systems. This being the case, if you define the grounding conductors as circuit conductors, there can never be separately derived systems.

 

[benray]

Thanks very much, Maybe my problem in understanding this issue is the NEC use of words not in the English language; ie, Grounded, and Grounding.

Very well composed response. Thanks again.

 

[busbar]

benray — That is an interesting review of electrical terminology changing over time, but with essentially unchanged meaning—partly stemming from the negotiated consensus of committees, I suppose.

I’ll try to address your very interesting comments all though the responder on this end is marginally qualified to do so. Though not necessarily accurate observations on my part—the “need” for autotransformer grounding is a regional {NEC} peculiarity and is not mandatory for successful operation unless—for pure economics—the bank has graded insulation like those typical of transmission voltages. It seems like the effectiveness {or tragedy} of system grounding has been hotly argued for at least a hundred years.

The two ‘ground’ connections on a ∆-Y set serve different purposes. On the hi-side, the connection to the ground source is usually a copper conductor, but could be metal raceway and still meet the intent of the US NEC. The hi-side equipment-grounding conductor comes into play for primary-to-ground faults, to operate the upstream overcurrent device(s), and to minimize potential difference between the transformer core/enclosure and adjacent grounded objects. On the other hand, no significant current flows in the primary equipment=-ground conductor for faults on ground faults, phase faults or overloads in secondary coils and the components they serve.

Although not directly/solidly grounded, 208>240V [only recently for other than ‘existing’ applications] and 480>600V applications are permitted by exception with an acceptable, safe, longstanding track record. Dime-a-dozen multi-tapped discharge-lighting ballasts operating consistently from 120, 208, 240 or 277 volts is one example of accommodation of systems—all on one continent.

The only real problem with autotransformer operation seems to be that the shunt winding must not be allowed to open circuit with current in the series winding. Else, the unit may experience core saturation and resultant insulation-damaging overvoltage.

Open-wye-secondary service is grounded by virtue of being fed universally from a 4-wire grounded-wye source. A dedicated open-wye arrangement serving any load is fairly rare, and would almost without exception be tapped from a 4-wire wye-secondary configuration. [Open delta may or may not be grounded based on local custom, particularly if the coil voltage is 480.]

Getting in over my head fast, Grd·Y—Grd·Y transformer banks variation in voltages passed from primary to secondary Grd·Y—Grd·Y will reflect positive, negative and zero-sequence components, where Ungrd·Y—Grd·Y connections will not pass zero-sequence {related to primary grounded neutral} voltages/currents. Ungrd·Y or Grd·Y choices in utility distribution seems to be a localized choice—there are arguments for both from an operational-safety versus economic view. Lineman discuss whether normally ungrounded bank-primary wyepoints should be temporarily grounded during fused-cutout switching.

 

[jghrist]

In my last post, I cavalierly stated that the secondary of a grd wye - grd wye transformer is not a separately derived system. I still believe this to be the case, but does this mean that you do not ground the neutral at the transformer and depend on the neutral ground back at the primary service point? Doesn't sound reasonable to me. Any AHJ's (Authorities Having Jurisdiction) out there want to answer this? Sounds like a good enough reason to me to not use this connection where you are subject to the NEC.

As far as the ungrd wye - grd wye connection, you don't really have an open in the zero-sequence circuit if the primary neutral is grounded somewhere, like back at the service. You just have an unknown ground resistance in the circuit. In a metal framed building, this might not be all that much resistance.